Download The Use of Monoclonal Antibodies and Flow

From www.bloodjournal.org by guest on August 10, 2017. For personal use only.
The Use of Monoclonal Antibodies and Flow Cytometry in the Diagnosis
Paroxysmal Nocturnal Hemoglobinuria
of
Bv Sharon E. Hall and Wendell F. Rosse
We have characterized the erythrocytes, granulocytes, and
platelets of 54 patients with paroxysmal nocturnal hemoglobinuria (PNH) with antibodies to glycosylphosphatidylinositol-anchored proteins (anti-CD55, anti-CD59, and anti-CD161
and flow cytometry to establish the usefulness of this technique in the diagnosis of this disorder. All patients demonstrated either completely (PNH 111) or partially (PNH I11
deficient red cells and granulocytes. Anti-CD59 best demonstrated PNH II red cells, which were present in 50% of the
patients. The proportion of abnormal granulocytes was usually greater than the proportion of abnormal red cells; 37%
of the patients had >80% abnormal granulocytes.Anti-CD55
did not delineate the erythrocyte populations as well as did
anti-CD59. Either anti-CD55 or anti-CD59 could be used
equally well toanalyze granulocytes; anti-CD16 did not demonstrate cells of partial deficiency. Platelets could not be
used fordetailed analysis asthe normal and abnormal populations werenotwell
distinguished. Flow cytometry of
erythrocytes using anti-CD59 or of granulocytes using either
anti-CD55 or anti-CD59 provides the most accurate technique for the diagnosis of paroxysmal nocturnal hemoglobinuria; it is clearly more specific, more quantitative, and
more sensitive than the tests for PNH that depend upon
hemolysis by complement (the acidified serum lysis [Ham]
test, the sucrose lysis test, and the complement lysis sensitivity [CLS] test).
0 1996 by The American Society of Hematology.
P
and unseparated granulocytes of S4 patients, purified granulocytes
of 39 patients,andplateletsof
18 patientswere tested by flow
cytometry.Samples of bloodfromhealthy adult volunteerswere
analyzed concurrently to determine normal values.
AROXYSMALnocturnalhemoglobinuria
(PNH) is
characterized by the deficiency, absolute or partial, of
all proteins anchored to the membrane by the glycosylphosphatidylinositol (GPI) anchor’”; at least 1S of these proteins
have been shown to be lacking on the abnormal blood cells
of patients with this disease. In the past, the diagnosis
of
PNH has relied on the demonstration
of the effect of the
absence o f t w oof these proteins, CD59 (membrane inhibitor
comof reactive lysis[MIRL]? protectin,s,6 membrane attack
plex inhibiting factor [MACIF],’ etc). and CDSS (decay accelerating factor [DAF],’ which results in the abnormal sensitivity of the red cells to the lytic action of complement.”“’
In the present study, we report the results of the use
of
monoclonal antibodies (MoAbs) against CDSS, CDS9, and
CD16 (the Fc,receptor 111) in the analysis byflow cytometry
of the red cells, granulocytes, and platelets
of 54 consecutive
PNH patients. We demonstrate the ability of this methodolof
ogy to distinguish the presence of even small populations
blood cells partially or totally deficient in these proteins and
to assess accurately the proportion of cells in these populawe establish for the first time the
tions. By this analysis,
optimum use of this methodology in the diagnosis of this
disease.
MATERIALS AND METHODS
Muterials
The MoAb 1H4, specific for CD55, was produced in our laboratory. An MoAb to CD59, 10G10, was the gift of Dr Marilyn Telen
(Duke University, Durham, NC). An MoAb to CD16 (3G8) was the
gift of Dr Mark Cunie. Mouse anti-human glycophorin A antibody
labeled withR-phycoerythrin(R-PE)waspurchasedfromPharmingen(SanDiego,CA).AnMoAb,CD13directly
labeled with
R-PEwasobtainedfromSouthernBiotechnologyAssociates,Inc
(Birmingham, AL). Nonreactive MoAb P3 with random specificity
was used as a negative control. Fluorescein-conjugated goat F(ab’
anti-mouse IgG(GAM-FITC) was purchased from Biosource International (Camarillo, CA). Normal goat serum and Dulbecco’s phosphate-bufferedsaline(DPBS)werepurchasedfromGIBCOBRL
(GrandIsland, NY). Humanserumwasobtainedfromnormal
healthy group AB volunteers. Ortho-mune lysing reagent was
purchased from Ortho Diagnostic Systems Inc (Raritan, NJ). Dextran
T500 was obtained from Pharmacia Biotech Inc (Piscataway,
NJ).
LymphocyteSeparationMedium(LSM)
was purchasedfrom Organon Teknika Corporation (Durham, NC). The buffer used in the
l b w cytometry assays was veronal-buffered saline (GVB-EDTA),
composed of 0.005 mol/L sodium barbital, 0.15 m o l L NaC1. 0.1o/r
gelatin, 0.01.5 mol/L EDTA, pH 7.5.
Patients
Preparatiort ?f Cells for Flow Cytometry
The blood of S4 patients (23 men, 3 I women) who had or were
suspected of having the diagnosis of PNH was tested. The patients
ranged in age from 1 1 to 65 years (mean 34 years). Erythrocytes
Venous peripheral blood was drawn from patients with PNH
or
suspected of having PNH and normal controls into EDTA-containing
tubes. For whole blood analysis, 400 p L of whole blood was removed and 100 pL was placed in each of four tubes. Two milliliterh
of Ortho-mune lysing reagent was added and incubated for 10 minutes. The tubes were centrifuged at 1500rpm for 5 minutes, and the
pellet washed once in GVB-EDTA. The pellet was
then resuspended
to 100 pL in GVB-EDTA. Erythrocytes were washed three times
to 1 X IO’/mLin GVB-EDTA.
in GVB-EDTAandresuspended
Granulocytes were purified from whole blood by diluting the blood
with an equal volume of DPBS and layering the mixture over an
equal amount of LSM. The mixture was spun at
2000 rpm for 30
minutes. The granulocytes were further purified
by sedimentation of
the erythrocytes with Dextran on ice for 45 minutes. The granulocytes were washed in GVB-EDTA and resuspended to 5 X 107/mL
in GVB-EDTA. Platelets were obtained by centrifuging whole
blood
From the Division of Hematology-Oncology,Department of Medicine, Duke University Medicul Center, Durham. NC.
Submitted June 9, 1995; accepted February 16, 1996.
Address reprint requests to Wendell F. Rosse, MD, Box 3934.
Duke University Medical Center, Durham, NC 27710.
The publication costs of this urticle were defrayed in part by page
charge puyment. This urticle must therefore be hereby marked
“advertisement” i n accordance nith I 8 U.S.C. section 1734 solelv to
indicate this fact.
0 1996 by The American Society of Hematologv.
0006-4971/96/8712-0002$3.00/0
5332
Blood, Vol 87,No 12 (June 15),
1996:
pp 5332-5340
From www.bloodjournal.org by guest on August 10, 2017. For personal use only.
5333
DIAGNOSIS OF PNH BY FLOW CYTOMETRY
at 800 rpm for 15 minutes and removing the platelet-rich plasma.
The platelet-rich plasma was then centrifuged at 1500 rpm for 5
minutes, and the platelet pellet was washed twice and resuspended
to 1 X 108/mLin GVB-EDTA.
Q
2 140
Immunophenotype Analysis
Q
Single color flow analysis was performed by using a two-step
method. One hundred microliters of the various cell suspensions
were incubated with 100 pL of each MoAb and P3 for 30 minutes
at room temperature. The cells were then washed twice with GVBEDTA and resuspended to 100 pL volume.A 10-pL mixture of
GAM-FITC, normal goat serum, and group AB normal human serum
(ratio 1:1:2.5) was added and incubated for another 30 minutes at
room temperature, followed by two washes in GVB-EDTA. The
pelleted cells were resuspended in 1 mL of buffer and analyzed on
an Ortho Cytoron Absolute (Ortho Diagnostic Systems Inc). Cell
types were determined by their forward and right scatter.
Two-color analysis was performed as above except that MoAbs
to specific cell markers and directly labeled with R-PE were used
at the recommended manufacturer’s dilutions as a third step. Color
compensation for crossover of fluorescence signals between the two
detectors was adjusted for optimal analysis.
Complement Lysis Sensitivity Test (CLS)
A modified version of the original CLS test of Rosse and Dacie’
was used to compare erythrocyte populations to flow cytometry. The
reaction volume was reduced to 375 pL withthe proportions of
erythrocytes, complement, and buffer staying the same. PNH cells
give complex curves that are analyzed to determine the three different PNH cell populations.
RESULTS
Erythrocytes
Normal erythrocytes. The erythrocytes from 44 normal
donors were examined with anti-CD55 and those of 49 normal donors with anti-CD59 (Table 1).The mean fluorescence
was greater with anti-CD59. Cells not reacting with the antibody used (unstained cells) were present in small numbers
in each case.
Erythrocytes from two normal donors were maintained at
4°C for up to 25 days to determine the stability of the detection of CD59. Six replicate samples were analyzed at each
point for percentage of cells bearing CD59 and for the mean
fluorescence obtained understandardized conditions. The
percentage of cells bearing CD59 remained stable in time
(range 98.3% to 99.8% and 98.4% to 99.7%, respectively)
Table 1. Analysis of Normal Erythrocytesby Flow Cytometry and
Anti-CD55 and Anti-CD59
~~
No. of samples
Mean fluorescence (channel)
Unstained cells
No. of samples’
Mean percentage
Range (%)
~
Anti-CD55
Anti-CD69
44
76.8 2 12.1
49
137.5 2 21.2
20 (45)
0.23 2 0.34
0-1.4
41 (83.7)
0.31 -c 0.39
0-1.6
* Values in parentheses are percentages.
0
g 120
v)
3
ii
c 100
m
80
l
0
5
I
1
,
l
~
15
20
Days in Storage
10
!
I
25
I
30
Fig 1. The mean fluorescence channel of normal cells (open symbols) and PNH I andlor PNH II cells (closed symbols) during storage
in plasma at 4°C.
as did the standard deviation
of the determination (range
0.03% to 0.94%). The meanfluorescencedecreasedwith
time (Fig 1).
PNH erythrocytes. The washed erythrocytes from 54 patients with the diagnosis of or suspected of having the diagnosis of PNH were examined by flow cytometry with antiCD59 and anti-CD%. Using anti-CD59, three types of cells
could be distinguished (Fig 2): cells with nearly normal expression of GPI-anchored protein (PNH I cells), cells with
no detectable expression of GPI-anchored proteins (PNH111
cells), and cells with intermediate expression of GPI-anchored proteins (PNH I1 cells). Cells of intermediate expression occurred either as a discrete population(Fig2c), as
“transition” between PNH 111 and PNH I cells (Fig 2d), or
as an indistinct major population (Fig 20. The mean fluorescence using anti-CD59 of these populations is shown in Fig
3. An intermediate population was more readily seen when
anti-CD59 was used, in part because the mean fluorescence
of the PNH I cells is greater than with anti-CD%.
The proportion of cells in each erythrocyte population was
analyzed and the results for anti-CD59 are shown in Table
2. The pattern of populations of PNH erythrocytes for 54
patients is shown in Table 3.
In 38 patients, the results of flow cytometry analysis were
compared to the results of the CLS test. The results were
concordant in 25 of 38 patients (65.8%) in that both identified the same populations, although the proportion of cells
in the populations may not have been entirely similar. The
CLS test did not detect a very small abnormal population of
cells in one patient, whereas no patients tested had abnormal
cells by CLS and none by flow cytometry. In the other 12
patients (31.2%), the CLS test did not identify correctly the
degree of abnormality (PNH I v PNH I1 v PNH 111) or did
not identifysmall populations (particularly small populations
of PNH I1 cells).
To determine the reproducibility and stability of the determinations of CD59 on PNH erythrocytes, the blood of three
patients was maintained at 4°C for up to 25 days and was
tested at intervals, analyzing six replicate samples at each
From www.bloodjournal.org by guest on August 10, 2017. For personal use only.
HALL AND ROSSE
5334
Normal
Fig 2. Histograms of fluorescence of erythrocytes using
anti-CD59. (a) Normal erythrocytes and inert antibody controls
(P3). (b through e) Representative patterns from selected patients. (b) PNH l and PNH 111 cells;
(c) discrete populations of PNH 1,
PNH II, and PNH 111 cells; (dl discrete PNH I and PNH II cells, indistinct (transition) PNH II cells;
(e) nearly entirely PNH 111 cells;
(f) primarily PNH II cells with indistinct PNH I and PNH 111.
.
point for percentage of abnormal (PNH I1 andor 111) cells
and the mean fluorescence channel of each population. The
fraction of abnormal cells diminished with time (Fig 4); the
mean fluorescence of PNH I and PNH I1 cells diminished at
the same rate as that of normal cells (Fig 1).
Analysis of Granulocytes
Normal granulocytes. The granulocytes, separated or
unseparated, from normal donors were examined with anti-
CD55 and anti-CD59 (Table 4). In many donors, unstained
cells were found; a larger proportion of these cells were
found when unseparated granulocytes were examined, particularly when anti-CD59 was used in analysis.
PNH granulocytes. The separated and unseparated granulocytes of 35 patients with PNH or suspected of having
PNH were analyzed using anti-CD%, anti-CD59, and antiCD16 (Fig 5a through f). The cells could be divided into
populations of PNH I, PNH 11, and PNH 111, using criteria
1;
20
.c,
15
c
.-a,
Y
I - 1 Il
L
e
3 5
Fig 3. The mean fluorescence
channel of anti-CD59 on erythronormal
erythrocytes. I:,
cytes !...; ,inert antibody(P3) controls; 0,PNH I cells; B
B
, PNH II
cells (discrete
population);
m,
PNH I1 cells (transition population); m, PNH 111 cells.
z
I
0
50
100
Mean Fluorescence Channel
I
I
-1
150
From www.bloodjournal.org by guest on August 10, 2017. For personal use only.
DIAGNOSIS OF PNH BY FLOW CYTOMETRY
5335
Table 2. Percentage of Abnormal Cells in Each of the Three PNH Populationsin Erythrocytes and Isolated Granulocytes
~________
Anti-CD551
PNH
Granulocytes
Percent
~
(Anti-CD59)
~~
Erythrocytes
I
0-20
20-40
40-60
60-80
80- 100
Total
9 (16.7)
15 (27.8)
8 ( 14.8)
15 (27.8)
7 (13.0)
54 (100)
PNH II
PNH Ill
PNH I
PNH II
PNH Ill
15 (26.8)
3 (5.4)
1 (1.8)
12 (21.4)
12 (21.4)
11 (19.6)
12 (21.4)
7 (12.5)
54 (96.4)
14 (40.0)
5 (14.3)
6 (17.1)
2 (5.7)
4 (11.4)
31 (88.6)
6 (17.1)
1 (2.9)
1 (2.9)
2 (5.7)
2 (5.7)
7 (20.0)
8 (22.9)
14 (40.0)
33 (94.3)
-
2 (3.6)
21 (37.5)
1 (2.9)
9 (25.7)
Values in parentheses are percentages.
similar to those used for analyzing red cells. The separation
of the populations and the ability to analyze accurately their
percentage was greater with separated than unseparated cells.
The mean fluorescence of the cells in each population is
shown for separated cells in Fig 6. The proportion of cells
in each population for 35 patients is shown in Table 2. The
pattern of populations of PNH granulocytes for 35 patients
is shown in Table 3.
The relationship between the proportion of cells in each
population to the proportion of cells in the corresponding
red cell population is shown in Fig 7. In 23 of 35 cases
(65.7%), the proportion of abnormal granulocytes was
greater than the proportion of abnormal red cells, in 7 cases
(20.0%) not significantly different, and in 5 cases (14.3%),
it was less.
Anti-CD16 also distinguished PNH granulocytes but did
not distinguish populations of intermediate content (PNH I1
cells).
Normal and PNH Platelets
Normal and PNH platelets were testedwith anti-CD55
and anti-CD59. The amount of both proteins detected on the
surface of normal cells was small with the result that the
populations of cells in PNH patients were not well distinguished.
DISCUSSION
by van der Schoot and associates in 1990 in the analysis of
granulocytes withapanel
of antibodies including antiCD16." These studies were confirmed by Schubert et all2;
in both studies, the totally deficient and normal populations
of cells were found in all blood elements but partially deficient cells were not identified. Red cells of intermediate
abnormality were identified in flow cytometry by Rosse et
all3 and Shichishima et al,'4s'5 and, in both studies, were
compared with the results obtained by the CLS test. More
recently, Kwong et all6 have indicated other antibodies that
may be used in detecting GPI-deficient cells in PNH. In none
of these studies were the antibodies used compared in a large
number of patients to establish the optimal conditions and
techniques that should be used in the diagnosis of PNH and
characterization of the cells in this disorder.
In the present study, we have analyzed the cells of 54
patients with PNH to establish the optimum conditions for
using flow cytometry in the diagnosis of the disorder. We
have found that the analysis of erythrocytes is best carried
out using anti-CD59. The mean fluorescence obtained with
the GPI-positive cells is sufficiently great as to be readily
distinguished from the GPI-negative cells; this was not the
case with anti-CD55 (anti-DAF). The mean fluorescence of
the GPI-positive PNH I cells in PNH was significantly less
50 I
I
T
The use of MoAbs to GPI-linked proteins and flow cytometry in the definition of the defect in PNH was firstproposed
Table 3. Patterns of Populations of Erythrocytes and Granulocytes
in the Blood of Patients with PNH
Populations*
Erythrocytes
(Anti-CD59)
Granulocytes
(Anti-CD55)
PNH I + II
PNH I + II + 111
PNH I + (11)
111
PNH I + 111
PNH II
PNH II + 111
PNH Ill
PNH (I + II) + 111
+
15 (27.8)
6 (11.1)
27 (50)
2 (3.6)
4 (7.1)
1 (2.9)
2 (5.7)
1 (2.9)
18 (51.4)
1 (2.9)
2 (5.7)
1 (2.9)
9 (25.7)
Values in parentheses are percentages.
* Parentheses indicate that the population is indistinct or transitional.
1
I"
0
5
10
15
Day
20
25
Fig 4. Changes in the percent of abnormal (PNH II [open rymbolsl
andlor PNH 111 [dosedsymbolsl) cells with time instorage in plasma
at 4°C. Each point is the mean of six replicate doterminatlons; l SD
is indicated by the bars associated with it.
From www.bloodjournal.org by guest on August 10, 2017. For personal use only.
HALL AND ROSSE
5336
Table 4. Analysis of Isolated and Nonisolated Normal Granulocytes by Flow Cytometry and Anti-CD55 and Anti-CD59
Separated'Granulocytes
AntiCD55
No. of samples
Mean fluorescence (channel)
Unstained cells
No. of samples+
Mean percentage
Range (%)
15
119.7
't
7.6
6 (40)
0.15 't 0.25
0-0.9
Anti-CD59
16
132.1 2 32
9 (56)
0.13 2 0.19
0-0.6
Unseparated Granulocytes
AntLCD55
50
136.5 't 10.7
47 (94)
0.49 't 0.74
0-3.6
Anti-CD59
53
160
z
39
51 (96.2)
2.89 't 4.12
0-24.6
Values in parentheses are percentages.
than that of normal cells, a result also obtained by Shichishima et al"; this suggests that the PNH I cells of PNH may
not be entirely normal in their content of GPI-linked proteins.
A very small number of cells not reacting with the antibodies was found in the blood of a few normal control subjects.
Analysis by two-color cytometry as well as repeated analyses
from the same patient suggested that these did not represent
a small population of PNH-like cells. Their presence did,
however, set the lower limit of detection of a PNH population at about 1% of the cells.
Red cells of intermediate expression of GPI-dependent
proteins were more readily identified using anti-CD59 than
anti-CD55 These appeared to occur in three patterns: a distinct population of cells (17 patients), a continuum of cells
between the PNH I and I11 cells (8 patients) and as a majority
of the cells with deficient and nearly normal cells at each
end of the spectrum (2 patients). All in all, 27 of 54 patients
were found to have cells of intermediate expression. These
findings correspond well with the findings using special tests
of complement sensitivity and cell separation." However,
when compared with the complement lysis sensitivity test
alone, PNH I1 cells weremorereadilyidentified
byflow
cytometry.
These studies demonstrate the degree of variability of the
determination of the fraction of PNH cells in each population
and the amount of CD59 on the membrane (by determination
of the mean fluorescence channel) by replicate sampling. In
addition, they show that the amount of CD59 onnormal
cells and PNH I and I1 cells gradually decreases as the cell
is stored; this protein may be digested by enzymes in the
plasma or may dissociate from the membrane, as has been
demonstrated for GPI-linked proteins." They also demonstrate that the fraction of abnormal cells consistently falls
over time; the reason for this is not apparent. These two
factors suggest that analysis is best performed with samples
as fresh as possible.
The analysis of granulocytes for the deficiency of GPIdependent proteins can also be used to diagnose PNH.".l6
AI1 three antibodies used (anti-CD59, anti-CD%, and antiCD16) readily demonstrated the deficiency of GPI-dependent proteins in some of the cells of all patients andthe
proportion detected by each was approximately the same.
Fig
5.
Histograms of fluorescence of granulocytesusing
anti-CD59 and FITC-labeled antiIgG. (a) Normal granulocytes; (b)
inert antibody (P3); IC)PNH I and
PNH 111 cells; Id) discrete populations of PNH 1, PNH It, and PNH
111 cells; (e) PNH 1 and PNHII cells;
(f) nearly entirely PNH 111 cells.
From www.bloodjournal.org by guest on August 10, 2017. For personal use only.
DIAGNOSIS
OF PNH BY FLOW CYTOMETRY
Fig 6. The mean fluorescence
channel ofanti-CD59 on granulocytes. [ : I ,
normal granulocytes; i.’. , inert antibody (P31
controls; 0,PNH I cells; B, PNH
It; W, PNH 111 cells.
5337
L
50
Mean Fluorescence Channel
The cells of intermediate abnormality (PNH I1 cells) were
more difficult to demonstrate than in red cells. PNH I1 cells
were not detected by anti-CD16, and the proportion of patients having PNH I1 granulocytes (demonstrated by antiCD55 or anti-CD59) was less than that having PNH I1 erythrocytes.
Two techniques were used to prepare the granulocytes:
the unseparatedgranulocyte technique, in which the red
8
8
8
8
8
m
8
:
m
~
“0
100
20
40
60
80
Percent Abnormal RBC’s
100
Fig 7. A comparison of the proportion of abnormal (PNH II + PNH
111 cells) erythrocytes and the proportion of abnormal granulocytes
in the blood of 35 patients with PNH.
cells of whole blood were hemolyzed and the granulocytes
reacted with the antibody and analyzed, and the separated
granulocyte technique, in which the granulocytes were separated from other blood elements by differential centrifugation before testing. Whereas abnormal granulocytes could
always be detected in the simpler unseparated assay, the
distinctions between different populations and the proportions of cells in each population were less clear than in the
tests using separated granulocytes. With both tests, a small
number of “granulocytes” in normalblooddidnot
stain
with the antibody; as with the red cells, this was not consistent in a given patient and was thought to be artifactual.
However, the number of cells in normal blood that did not
bind the antibody was very muchhigher with the unseparated
granulocyte technique; for this reason, that technique, although simpler, should be used with caution in the analysis
of blood for the diagnosis of PNH, particularly whenthe
proportion of abnormal cells is low. Samples older than 2
days could not be used for analysis because of poor quality
of the cells.
In general, the proportion of abnormal granulocytes detected was greater than the number of abnormal red cells
from the same blood sample. This observation has been previously found by flow cytometry and other techniques.I8The
explanation lies in part in the fact that the survival in the
circulation of the abnormal granulocytes appears to be normal,’’ whereas that of the abnormal red cells is reduced.2”
Thus, the percentage of circulating abnormal granulocytes
more nearly represents the proportion of abnormal cells delivered to the circulation.
Cells with less thannormal amounts of the GPI-linked
proteins (PNH I1 cells) were demonstrated in both erythrocytes and granulocytes in a majority of cases when anti-
From www.bloodjournal.org by guest on August 10, 2017. For personal use only.
5338
HALL AND ROSSE
CD59 was used for detection. These cells were less easily
seen with anti-CD55 and were not observed in granulocytes
with anti-CD16. In some cases, the partial deficiency is due
to a missense mutation of the pig-A gene2',22;in others, the
identified defect appears to be one thatwouldleadto
a
premature stop codon.22The reason for this discrepancy is
under investigation.
The deficiency of the GPI-dependent proteins couldbe
demonstrated in the platelets of patients with PNH but the
fluorescence of normal and PNH I platelets was little different from the negative control and the distinction between
PNH I and PNH I11 platelets was difficult and imprecise.
GPI-dependent protein deficient lymphocytes are demonstrable in most ifall patients
not
with PNH'Z.'h,23-26.
,Immortalized lymphocyte cell lines have played an important role in
the identification of the genetic defect in PNH.27-2y
In most
patients, the proportion of abnormal lymphocytes is small,
particularly relative to the proportion of abnormal erythrocytes or granulocytes in the same blood. In some studies,
the abnormal lymphocytes were missed entirely if they were
in fact present.'' Thus, analysis of lymphocytes for GPIlinked proteins to make the diagnosis of PNH is not recommended. On the other hand, in patients who recover from
PNH with disappearance of abnormal erythrocytes and granulocytes may have a persistent population of GPI-deficient
lymphocytes." This is thought to be due to the longevity of
lymphocytes compared with other blood cells.
How cytometry andMoAbsto GPI-dependent proteins
have been used to analyze the cells of the bone marrow in
PNH. In general, the proportion of abnormal cells corresponds roughly to that of the granulocytes. In exceptional
cases, the abnormal cells in the bone marrow can bedetected
in the marrow for several months before they appear in the
peripheral blood.3'
Small populations of cells deficient in GPI-dependent proteins can be demonstrated in the blood of 15% to 30% of
patients with aplastic anemia, particularly after treatment
with antithymocyte
These populations are diagnostic of PNH by definition, but the clinical syndrome often
remains that of aplastic anemia. These findings indicate a
close but not clearly understood relationship between the
two ~yndromes.'~,~'
These studies clearly demonstrate the utility of MoAbs
and flow cytometry in the diagnosis of paroxysmal nocturnal
hemoglobinuria. Flow cytometry is shown in these studies
to be superior to the CLS assay in that small populations of
abnormal cells may be missedoccasionally in the CLS test (1
patient in this series). Further, flow cytometry distinguishes
much better the cells of intermediate abnormality (PNH I1
cells); while this may not be of great interest diagnostically,
it is of major importance in characterizing the abnormalities
in the cells of different patients.
We have previously shown that the CLS test was superior
to either the Ham test or the sucrose lysis test?' Even when
the Hamtest is done well, it does not detect small populations
of abnormal cells (which results in a decrease in sensitivity),
the degree of lysis does not accurately reflect the proportion
of abnormal cells,"' and the degree of abnormality of the
'
abnormal cells cannot be assessed. Furthermore, the test is
falsely positive in the rare disorder hereditary erythrocytic
multinuclearity with a positive acidified-serum lysistest
([HEMPAS] congenital dyserythropoietic anemia, type H).'''
The sucrose lysis test is technically easier to performbut
has a higher rate of false-positive resultsu (thus decreasing
its specificity). Although it more accurately detects the proportion of PNH 111 red cells than the Ham test, it likewise
does not delineate the abnormality of these cells and does
not quantitatively delineate the number of PNH I1 cells."'
Flow cytometry has, in addition, the advantage that the
reagents can be standardized; the tests depending upon hemolysis use normal human serum as a reagent and there may
be great variability in the potency of normal serum to lyse
the abnormal red cells.'s Further, the reproducibility of the
results from flow cytometry is greater thanthat of either
other test. Red cells can be stored up to 3 weeks without
change in the characteristics of the cells whereas granulocytes are difficult to analyze if they are more than 12 to 24
hours old. The estimated costs of this more accurate diagnostic technique may be 1 to 1.5 times as much as the conventional but less accurate serological tests.
Flow cytometry in the diagnosis of PNH is specific in that
no other clinical syndrome is characterized by cells with the
phenotype demonstrated by this technique. Hematopoietic
cells congenitally deficient inCD5546.47and CD594Rhave
been described, but these extremely rare conditions are
readily distinguished from PNH by the facts that 100% of
the cells are deficient in these syndromes and only one of
the proteins is deficient. Because flow cytometry is able to
demonstrate the characteristic cells with greater certainty and
precision than previous methods using secondary effects (eg,
increased sensitivity to complement lysis), we propose that
the analysis byflow cytometry of erythrocytes using antiCD59 or of granulocytes using either anti-CD59 or antiCD55 be the standard diagnostic test for paroxysmal nocturnal hemoglobinuria.
REFERENCES
1. Rosse WF: Phosphatidylinositol-linked proteins and paroxysmal nocturnal hemoglobinuria. Blood 75: 1595, 1990
2. Yeh ETH, Rosse WF: Paroxysmal nocturnal hemoglobinuria
and the glycosylphosphatidylinositol anchor. J Clin Invest 93:2305,
1994
3. Takeda J, Miyata T, Kawagoe K, Iida Y , Endo Y, Fujita T,
Takahashi M, Kitani T, Kinoshita T: Deficiency of the GP1 anchor
caused by a somatic mutation of thePIG-A gene in paroxysmal
nocturnal hemoglobinuria. Cell 73:703, 1993
4. Holguin MH, Frederick LR, Bernshaw NJ, Wilcox LA, Parker
CJ: Isolation and characterization of a membrane protein from normal human erythrocytes that inhibits reactive lysis of the erythrocytes of paroxysmal nocturnal hemoglobinuria. J Clin Invest 84:7,
1989
5. Men S, Waldmann H, Lachmann PJ: Distribution of protectin
(CD59), a complement membrane attack inhibitor, in normal human
tissues. Lab Invest 65532, 1991
6. Davies A, Simmons DL, Hale G , Harrison RA, Tighe H, Lachmann PJ, Waldmann H: CD59, an LY-6-like protein expressed in
human lymphoid cells, regulates the action of the complement mem-
From www.bloodjournal.org by guest on August 10, 2017. For personal use only.
DIAGNOSIS OF PNH BY FLOW CMOMETRY
brane attack complex on homologous cells. J Exp Med 170:637,
1989
7. Sugita Y, Nakano Y, Oda E, Noda K, Tobe T, Miura NH,
Tomita M . Determination of carboxyl-terminal residue and disulfide
bonds of MACE (CD59), a glycosyl-phosphatidylinositol-anchored
membrane protein. J Biochem (Tokyo) 114:473, 1993
8. Nicholson-Weller A, March JP, Rosenfeld SI, Austen KF: Affected erythrocytes of patients with paroxysmal nocturnal hemoglobinuria are deficient in the complement regulatory protein, decay
accelerating factor. Proc Natl Acad Sci USA 805430, 1983
9. Rosse WF, Dacie JV: Immune lysis of normal human and
paroxysmal nocturnal hemoglobinuria red blood cells. I. The sensitivity of PNH red cells to lysis by complement and specific antibody.
J Clin Invest 45:736, 1966
10. Rosse WF, Adams JP, Thofpe AM: The population of cells
in paroxysmal nocturnal haemoglobinuria of intermediate sensitivity
to complement lysis: Significance and mechanism of increased immune lysis. Br J Haematol 28:181, 1974
11. van der Schoot CE, Huizinga TW, van't Veer-Korthof ET,
Wijmans R, Pinkster J, von dem Borne AE: Deficiency of glycosylphosphatidylinositol-linked membrane glycoproteins of leukocytes
in paroxysmal nocturnal hemoglobinuria; description of a new diagnostic cytofluorometric assay. Blood 76:1853, 1990
12. Schubert J, Alvarado M, Uciechowski P, Zielinska Skowronek M, Freund M, Vogt H, Schmidt RE: Diagnosis of paroxysmal
nocturnal haemoglobinuria using immunophenotyping of peripheral
blood cells. Br J Haematol 79:487, 1991
13. Rosse WF, Hoffman S, Campbell M, Borowitz M, Moore JO,
Parker CJ: The erythrocytes in paroxysmal nocturnal haemoglobinuria of intermediate sensitivity to complement lysis. Br J Haemato1 79:99, 1991
14. Shichishima T, Terasawa T, Hashimoto C, Ohto H, Uchida
T, Maruyama Y: Heterogenous expression of decay accelerating
factor and CD59/membrane attack complex inhibition factor on paroxysmal nocturnal haemoglobinuria (PNH) erythrocytes. Br J
Haematol 78:545, 1991
15. Shichishima T, Terasawa T, Saitoh Y, Hashimoto C, Ohto
H, Maruyama A: Diagnosis of paroxysmal nocturnal haemoglobinuria by phenotypic analysis of erythrocytes using two-color flow
cytometry with monoclonal antibodies to DAF and CD59IMACIF.
Br J Haematol 85:378, 1993
16. Kwong YL, Lee CP, Chan TK, Chan LC: Flow cytometric
measurement of glycosylphosphatidyl-inositol-linked surface proteins on blood cells of patients with paroxysmal nocturnal hemoglobinuria. Am J Clin Pathol 102:30, 1994
17. Kooyman D, Byrne G, McClellan S, Nielsen D, Kagan D,
Coffman T, Masahide T, Waldmann H, Platt J, Logan J: Erythroidspecific expression of human CD59 and transfer to vascular endothelial cells. Transplant Proc 26:1241, 1994
18. Stem M, Rosse WF: Two populations of granulocytes in paroxysmal nocturnal hemoglobinuria. Blood 53:928, 1979
19. Brubaker L, Essig LJ, Mengel CE: Neutrophil life span in
paroxysmal nocturnal hemoglobinuria. Blood 50:657, 1977
20. Rosse W F : The life-span of complement-sensitive and -insensitive red cells in paroxysmal nocturnal hemoglobinuria. Blood
37:556, 1971
21. Ware RE, Rosse WF, Howard TA: Mutations within the Piga gene in patients with paroxysmal nocturnal hemoglobinuria. Blood
839418, 1994
22. Bessler M, Mason PJ, Hillmen P, Luzzatto L: Mutations in
the PIG-A gene causing partial deficiency of GPI-linked surface
proteins (PNH II)in patients with paroxysmal nocturnal haemoglobinuria. Br J Haematol 87:863, 1994
23. Schubert J, Uciechowski P, Delany P, Tischler HJ, Kolanus
5339
W, Schmidt RE: The PIG-anchoring defect in NK lymphocytes of
PNH patients. Blood 761181, 1990
24. Tseng JE,Hall SE, Ware RE: Phenotypic and functional analysis of lymphocytes in paroxysmal nocturnal hemoglobinuria. Am
J Hematol 50:244, 1995
25. Masuda T, Yonoemura Y, Fujimoto K, Hidaka M, Nagakura
S , Nakakuma H, Hata H, Samada 0, Takatsuki T Establishment of
a humanT-cell line with deficient surface expression of glycosylphosphatidylinositol (GP1)-anchored proteins from a patient with paroxysmal nocturnal haemoglobinuria. Br J Haematol 87:24, 1994
26. Nakakuma H, Nagakura S, Horikawa K, Hidaka M, Kawaguchi T, Iwamoto N, Sanada I, Kagimoto T, Takatsuki K Interleukin-2-dependent T-cell lines established from paroxysmal nocturnal
hemoglobinuria patients. Blood 84309, 1994
27. Hillmen P, Bessler M, Mason PJ, Watkins WM, Luzzatto L:
Specific defect in N-acetylglucosamine incorporation in the biosynthesis of the glycosylphosphatidylinositol anchor in cloned cell lines
from patients with paroxysmal nocturnal hemoglobinuria. Proc Natl
Acad Sci USA 90:5272, 1993
28. Hillmen P, Bessler M, Crawford DH, Luzzatto L Production
and characterization of lymphoblastoid cell lines with the paroxysmal nocturnal hemoglobinuria phenotype. Blood 81:193, 1993
29. Bessler M, Mason PJ. Hillmen P, Miyata T, Yameda N,
Takeda J, Luzzatto L, Kinoshita T: Paroxysmal nocturnal hemoglobinuria (PNH) is caused by somatic mutations in the PIG-A gene.
EMBO J 13:110, 1994
30. Hillmen P, Lewis SM, Bessler M, Luzzatto L, Dacie JV:
Natural history of paroxysmal nocturnal hemoglobinuria. N Engl J
Med 333:1253, 1995
31. Nakakuma H, Nagakura S, Iwamoto N. Kawaguchi T, Hidaka
M, Horikawa K, Kagimoto T, Shido T, Takatsuki K: Paroxysmal
nocturnal hemoglobinuria clone inbonemarrowof
patients with
pancytopenia. Blood 85:1371, 1995
32. de Planque MM, Bacigalupo A, Wursch A, Hows JM, Devergie A, Frickhofen N, Brand A, Nissen C: Long-term follow-up of
severe aplastic anaemia patients treated with antithymocyte globulin.
Severe Aplastic Anaemia Working Party of the European Cooperative Group for Bone Marrow Transplantation (EBMT). Br J Haemato1 73:121, 1989
33. Antin JR. Ginsburg D, Smith BR, Nathan D C , Orkin SH,
Rappeport JM: Bone marrow transplantation for paroxysmal nocturnal hemoglobinuria: Eradication of the PNH clone and documentation of complete lymphohematopoietic engraftment. Blood 66:1247,
1985
34. Kruatrachue M, Na-Nakorn S : Transient paroxysmal nocturnal hemoglobinuria during the course of aplastic anemia. J Med
Assoc Thai 57:427, 1974
35. Schubert J, Vogt HG, Zielinska-Skowronek M, Freund M,
Kaltwasser JP, Hoelzer D, Schmidt RE: Development of the glycosylphosphatidylinositol-anchoringdefect characteristic for paroxysmal nocturnal hemoglobinuria in patients with aplastic anemia.
Blood 83:2323, 1994
36. Nissen C, Moser Y, dalle Carbonare V, Gratwohl A, Speck
B: Complete recovery of marrow function after treatment with antilymphocyte globulin is associated with high, whereas early failure
and development of paroxysmal nocturnal haemoglobinuria are associated with low endogenous G-CSA-release. Br J Haematol72:573,
1989
37. Najean Y, Haguenauer 0:Long-term (5 to 20 years) evolution
of nongrafted aplastic anemias. The Cooperative Group for theStudy
of Aplastic and Refractory Anemias. Blood 76:2222, 1990
38. Fujioka S, Yamada T: Decay accelerating factor and CD59
expression in peripheral blood cells in aplastic anaemia and report
From www.bloodjournal.org by guest on August 10, 2017. For personal use only.
5340
of a case of paroxysmal nocturnal haemoglobinuria secondary to
aplastic anaemia. Br J Haematol 83:660, 1993
39. Dacie JV, Lewis SM: Paroxysmal nocturnal haemoglobinuria:
Variation in clinical severity and association with bone marrow hypoplasia. Br J Haematol 7:442, 1961
40. Lewis SM, Dacie JV: The aplastic anaemia-paroxysmal nocturnal haemoglobinuria syndrome. Br J Haematol 13:236, 1967
41. Rosse WF: Paroxysmal nocturnal hemoglobinuria, in Rosse
WF (ed): Clinical Immunohematology: Basic Concepts and Clinical Applications, Boston, Blackwell Scientific Publications, 1990,
p 593
42. May JE, Rosse W, Frank MM: Paroxysmal nocturnal hemoglobinuria. Alternate-complement-pathway-mediatedlysis induced
by magnesium. N Engl J Med 289:705, 1973
43. Crookston JH, Crookston MC, Rosse WF: Red-cell abnormalities in HEMPAS (hereditary erythroblastic multinuclearity
with a positive acidified-serum test). Br J Haematol 23:83, 1972
(SUPPI)
HALL AND ROSSE
44. Hartmann RC, Jenkins DEj: The “sugar water” test for paroxysmal nocturnal hemoglobinuria. New Engl J Med 275: 155, 1996
45. Sirchia G, Ferrone S, Mercuriali F: In vitro lysis of AETtreated normal red cells (PNH-like cells) by normaland cirrhotic
sera. Brit J Haematol 17:245, 1969
46. Reid ME, Mallinson G, Sim RB, Poole J, Pausch V, Merry
AH, Liew YW, Tanner MJ: Biochemical studies on red blood cells
from a patient with the Inab phenotype (decay-accelerating factor
deficiency). Blood 78:3291, 1991
47. Tate CG, Uchikawa M, Tanner MJ, Judson PA, Parsons SF,
Mallinson G, Anstee DJ: Studies on the defect which causes absence
of decay accelerating factor (DAF) from the peripheral blood cells
of an individual with the Inab phenotype. Biochem J 261:489, 1989
48. Yamashina M,Ueda E, Kinoshita T, Takami T, Ojima A,
Ono H, Tanaka H, Kondo N, Orii T, Okada N, Okada H, Inoue K,
Kitani T: Inherited complete deficiency of 20-kilodalton homologous
restriction factor (CD59) as a cause of paroxysmal nocturnal hemoglobinuria. N Engl J Med 323:1184, 1990
From www.bloodjournal.org by guest on August 10, 2017. For personal use only.
1996 87: 5332-5340
The use of monoclonal antibodies and flow cytometry in the
diagnosis of paroxysmal nocturnal hemoglobinuria
SE Hall and WF Rosse
Updated information and services can be found at:
http://www.bloodjournal.org/content/87/12/5332.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.