Univ of PENN Health System PENN Flow Lab Home Page


John C. Pui, M.D., Richard Schreitzenmeir, B.S., and Jonni S. Moore, Ph.D.

Clinical Flow Cytometry Laboratory, Department of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA

PENN Flow Lab Home Page


Diagnostic Tests: Flow Cytometry
Cell Lineages
Antigen Choice
Clinical Examples
Aplastic Anemia


Paroxysmal nocturnal hemoglobinuria is a rare acquired clonal hematopoietic stem cell defect with an estimated frequency of 1-10 per one million [1].

PNH patients have an acquired somatic mutation in their PIG-Agene, located on the X-chromosome. The PIG-A gene codes for an as yet unidentified protein that is necessary for the addition of N-acetylglucosamine (GlcNAc) to PI as the first step to synthesizing the glycan core linkage between PI and the protein. Affected PNH cells cannot synthesize this glycan core, resulting in absent or diminished expression of GPI-linked proteins on the cell membrane [2]. Two GPI-linked proteins that are linked to increased RBC sensitivity to complement lysis and in vivo hemolysis are decay-accelerating factor (DAF, CD55) and membrane inhibitor of reactive lysis (MIRL, HRF, CD59). These two proteins are responsible for the regulation of the complement cascade. DAF accelerates the natural decay of the C3-C5 convertases. Absence of DAF alone, seen in the blood group phenotype Inab,results in only slightly increased in vitro complement sensitivity and no in vivo hemolysis. MIRL regulates the olymerization of the C9 membrane attack complex. Absence of MIRL alone, seen in extremely rare congenital deficiencies, does result in marked in vitro sensitivity of RBCs to complement lysis, and these patients have clinically significant hemolytic episodes similar to PNH patients [2].

The 10 year actuarial survival is between 50-70%, with the major complications and predictors of poor prognosis being thrombosis, pancytopenia, and myelodysplastic syndrome [1, 3].


Several tests exist for PNH, including those based on complement-mediated lysis (sugar water test, Ham's test, complement lysis sensitivity test; however, the sensitivity of these tests in the detection of small populations has recently been questioned [4, 5]. Several flow cytometric tests have been proposed, based on detection of alterations in GPI-linked surface receptors indifferent hematopoietic populations [4, 6, 7, 8, 9]. We investigated the relative merits of the various protocols and determined a recommended strategy for sensitive flow cytometric detection of PNH clones.

Cell Lineages

It has been shown that in the majority of cases, patients with PNH clones have erythrocyte, granulocyte, and monocyte lineages affected. Occasionally, PNH clones can be detected by flow cytometry in only the RBCs, or in the granulocytes and monocytes. In a minority of cases, the lymphocyte lineage is also affected, and only a few rare case reports have documented the lymphocytes to be the only cell lineage affected [3]. The prevalence of affected platelets in PNH has not been studied.

Examination of erythrocytes (RBC) can be very sensitive. The results of flow cytometry on RBCs can be compared with other testing techniques for validation, including the Ham test, which, as mentioned above, has traditionally been the gold standard for PNH diagnosis[9]. The red cell antigens are relatively stable so whole blood samples can be stored for up to 3 weeks at 4°C before analysis; however, there is a small decrease in fluorescent intensity, as well as a decrease in the proportion of affected RBCs over time, so that smaller populations of affected cells may be lost to detection [4]. Flow cytometry on RBCs alone may also underestimate extent of the PNH clone because of decreased PNH RBC survival in vivo, with subsequent enrichment of the sample for RBCs with normal GPI-linked protein expression, especially if the sample is obtained shortly after a hemolytic episode. Additionally, PNH patients who are anemic and have been recently transfused may have analysis complicated by the transfused RBCs.

Discrimination between positive and negative populations is also possible by analysis of granulocytes. Usually there are sufficient numbers of circulating granulocytes for analysis, and since affected cells have normal in vivo survival, there is no problem with enrichment of normal cells [4]. Analysis of granulocytes can also bypass the problem of transfused patients, especially if the transfused units were filtered or prestorage leukocyte depleted. A limitation of granulocyte analysis is that samples more than two days old cannot be reliably analyzed[4].

While monocytes also show good separation between positive and negative populations, there are usually smaller numbers in peripheral blood compared to granulocytes[4, 6, 7]. This could prove to be a problem in analyzing pancytopenic patients. There have been no studies regarding the reliability of monocyte analysis for PNH following a period of storage.

Since normal patients already have populations of lymphocytes that have weak expression of GPI-linked proteins, differential expression of these antigens is not as useful for PNH diagnosis[10].

Platelets have not been extensively studied because the amount of antigen on the platelets is relatively small so that there is difficulty in resolving positive events from negative events [7].

Antigen Choice

Many different GPI-linked protein antigens have been studied in PNH. These include:

Cell Lineage
LPS receptor
FcIII receptor
B-cell differentiation marker
all lineages
all lineages
possible adhesin
all lineages
CEA-related glycoprotein

One study evaluating the above antigens on various cell lineages showed that the best antigen for RBCs in the diagnosis of PNH was CD59, the best for granulocytes were CD66b, CD59, and CD55, and the best for monocytes were CD14 and CD55 [6].

CD55 and CD59 have been the antigens that have been the most intensively studied [4, 6, 7, 9]. Anti-CD59 analysis of RBCs is able to separate samples from PNH patients in three groups, one with close to normal expression of CD59, a second with diminished expression of CD59, and the third with absent expression of CD59, similar to type I, type II, and type III RBCs as defined by the CLS test [4].

Anti-CD55 analysis of granulocytes yield results similar to those seen in RBCs [4, 7]. The flow analysis correlates well with the CLS test for division of RBCs into type I, type II, and type III cells. Additionally, in cases where the RBC type II population is very small, flow cytometry is more sensitive in detecting these populations than the CLS test [9].

The majority of studies have used single-parameter analysis for looking at CD55 and CD59 expression. Only one Japanese study used two-parameter analysis of RBCs for CD55 and CD59 expression[9]. While single-parameter analysis of RBCs and granulocytes is usually able to distinguish the different cell populations, multiparameter analysis can provide enhanced resolution when there may be overlap between populations, especially in distinguishing type II and type III cell populations. Multiparameter analysis can also add to specificity in the instance of patients with isolated deficiencies of either CD55 (Inab phenotype) or CD59 (congenital 20kdHRF deficiency).

Clinical Examples


Because of the well known association between aplastic anemia (AA) and PNH, flow cytometry studies have also been performed on AA patients to look for the presence of a PNH clone [5, 11, 12]. A statistically significant association between clinical response to therapy for AA and the presence of a PNH clone has been reported in patients with a poorer response to therapy [5]. This suggests a potential usage of this type of analysis in real-time patient monitoring of therapy.


This test is more sensitive and specific in detecting the presence of circulating cells with deficient or absent expression of GPI-linked proteins than the traditional gold standard Ham's test. In the clinical setting of a patient with intravascular hemolysis, unexplained deep venous thrombosis, or pancytopenia,it is useful for the diagnosis of PNH. Because of its increased sensitivity and ability to analyze multiple cell lineages, it can detect the presence of PNH clones in AA patients in the absence of RBC involvement.

Based on our review of the literature and our clinical experience, our current recommendations for diagnosis of PNH via flow cytometry are as follows:

The specimen should consist of 5 mL of EDTA-anticoagulated whole blood.

If there is going to be a delay of more than 6 hours, the specimen should be stored at 4°C.

The RBCs are directly stained with two-color CD55 and CD59.

The granulocytes and monocytes are directly stained, using a whole blood lysis method and analyzed with three colors. Granulocytes will be stained with CD55, CD59, and CD66b; monocytes will be stained with CD55, CD59, and CD14.

Separation and gating of the RBC, granulocyte, and monocyte populations can be performed using forward scatter versus side scatter analysis.

The RBCs, granulocytes, and monocytes are analyzed using two-parameter CD55 vs. CD59 plots.

The granulocytes are also analyzed using single-parameter CD66b histograms.

The monocytes are also analyzed using single-parameter CD14 histograms.

Lymphocytes and platelets will not be analyzed.

A normal control is run in parallel to the specimen as a positive control (Type I cells).

Negative controls are determined from the isotype control (Type III cells).

A total of 10000 events are collected. The lower limit of detection in our laboratory is approximately 2-5%.

Patient samples that demonstrate cell populations with diminished or absent GPI-linked proteins (Type II or III cells) with multiple antibodies are considered to be consistent with PNH. More studies are needed to better define whether the type (I, II, or III), cell lineage, and size of the circulating clone can provide additional prognostic information.


1. Hillmen P, Lewis SM, Bessler M, Luzzatto L, Dacie JV. Natural history of paroxysmal nocturnal hemoglobinuria. New England Journal of Medicine. 333(19):1253-8, 1995, Nov 9.

2. Rosse WF, Ware RE. The molecular basis of paroxysmal nocturnal hemoglobinuria. Blood. 86(9):3277-86, 1995, Nov 1.

3. Socie G, Mary JY, deGramont A, Rio B, Leporrier M, Rose C, Heudier P, Rochant H, Cahn JY, Gluckman E. Paroxysmal nocturnal haemoglobinuria: long-term follow-up and prognostic factors. French Society of Haematology. Lancet. 348(9027):573-7, 1996 Aug 31.

4. Hall SE, Rosse WF. The use of monoclonal antibodies and flow cytometry in the diagnosis of paroxysmal nocturnal hemoglobinuria. Blood. 87(12):5332-40, 1996, Jun 15.

5. Schubert J, Vogt HG, Zielinska-Skowronek M, Freund M, Kaltwasser JP, Hoelzer D, Schmodt RE. Development of the glycosylphosphatidylinositol-anchoring defect characteristic for paroxysmal nocturnal hemoglobinuria in patients with aplastic anemia. Blood. 83(8):2323-8, 1994, Apr 15.

6. Alfinito F, Del Vecchio L, Rocco S, Boccuni P, Musto P, Rotoli B. Blood cell flow cytometry in paroxysmal nocturnal hemoglobinuria: a tool for measuring the extent of the PNH clone. Leukemia. 10(8):1326-30, 1996 Aug.

7. Fujioka S, Yamada T. Varying populations of CD59-negative, partially positive, and normally positive blood cells in different cell lineages in peripheral blood of paroxysmal nocturnal hemoglobinuria patients. American Journal of Hematology. 45(2):122-7, 1994 Feb.

8. Kwong YL, Lee CP, Chan TK, Chan LC. Flow cytometric measurement of glycosylphophatidyl -inositol-linked surface proteins on blood cells of patients with paroxysmal nocturnal hemoglobinuria. American Journal of Clinical Pathology. 102(1):30-5, 1994, Jul.

9. Shichishima T, Terasawa T, Saitoh Y, Hashimoto C, Ohto H, Maruyama Y. Diagnosis of paroxysmal nocturnal haemoglobinuria by phenotypic analysis of erythrocytes using two-colour flow cytometry with monoclonal antibodies to DAF and CD59/MACIF. British Journal of Haematology. 85(2):378-86, 1993, Oct.

10. Nagakura S, Nakakuma H, Horikawa K, Hidaka M, Kagimoto T, Kawakita M, Tomita M, Takatsuki K. Expression of decay-accelerating factor and CD59 in lymphocyte subsets of healthy patients and paroxysmal nocturnal hemoglobinuria patients. AmericanJournal of Hematology. 43(1):14-8, 1993, May.

11. Fores R, Alcocer M, Diez-Martin JL, Fernandez MN. Flow cytometric analysis of decay-accelerating factor (CD55) on neutrophils from aplastic anaemia patients. British Journal of Haematology. 90:728-30, 1995.

12. Griscelli-Bennaceur A, Gluckman E, Scrobohaci ML, Jonveaux P, Vu T, Bazarbachi A, Carosella ED, Sigaux F, Socie G. Aplastic

Back to Clinical Evaluation
Home Page Table of Contents Sponsors E-Mail Archive Web Sites

CD-ROM Vol 3 was produced by Monica M. Shively and other staff at the Purdue University Cytometry Laboratories and distributed free of charge as an educational service to the cytometry community. If you have any comments please direct them to Dr. J. Paul Robinson, Professor & Director, PUCL, Purdue University, West Lafayette, IN 47907. Phone:(765) 494-0757; FAX (765) 494-0517; Web http://www.cyto.purdue.edu, EMAIL cdrom3@flowcyt.cyto.purdue.edu