Paroxysmal Nocturnal Hemoglobinuria (PNH) is a rare, acquired stem cell disorder caused by a somatic mutation in the X-linked PIG-A gene ([1-4]). Depending on whether this mutation is a mis-sense or nonsense variant (or both), there is a partial or absolute defect in the biosynthesis and expression of glyco-phosphatidyl inositol- (GPI) linked structures including the complement-defense structures CD55 and CD59 ([5-8]). Absence of CD59 in particular ([9, 10]) and CD55 on red cells is largely responsible for intravascular hemolysis associated with clinical PNH (reviewed in ).
Since the early 1990s, flow cytometry to detect GPI-deficient cells has been the method of choice to rapidly diagnose PNH ([12-15]). Early methods relied upon the loss of CD55 and CD59 on red blood cells and granulocytes ([12, 13]), but such approaches were neither sufficiently accurate nor sensitive to detect small PNH clones (<1%) in aplastic anemia (AA) and myelodysplastic syndrome (MDS) cases. While a variety of more sensitive approaches have recently been developed for PNH white blood cell (WBC) detection based on a fluorescent derivative of Pro-aerolysin, FLAER ([16-20]), FLAER-based assays are not yet universally deployed and recent data from external quality assurance (EQA) programs have highlighted the variable capabilities of laboratories to accurately detect WBC PNH clones in stabilized whole blood samples ([21, 22]).
To address these issues, the International Clinical Cytometry Society (ICCS) published guidelines for the diagnosis and monitoring of PNH and related disorders by flow cytometry (). In order to improve PNH WBC detection, the ICCS Guidelines recommended the use of one lineage-gating antibody together with two GPI-linked reagents, one of which should preferably be FLAER. In developing a more detailed ‘Practical Guideline’, we detailed a 4-color combination for high-resolution detection of PNH granulocytes using FLAER, CD24, CD15 and CD45 that is highly sensitive and that can be deployed on a variety of clinical cytometers (). Similarly, a 4-color combination of FLAER, CD14, CD64 and CD45 was developed for high-resolution detection of PNH monocytes (). The granulocyte assay can reliably detect PNH phenotypes at the 0.02% level or higher, while the monocyte assay can reliably detect PNH phenotypes at ≥ 0.04% (). Background rates for the granulocyte and monocyte assays were assessed in 10 normal samples and found to be <0.0013% and <0.0033% respectively ().
The expression of variety of other GPI-linked structures on various blood cell lineages has been documented in peripheral blood ([25, 26]) and bone marrow (). Of these, CD66b was extensively deployed in the detection of PNH granulocytes ([21, 28]). However, until recently, CD66b was available only in FITC-conjugated form, ruling out its use in FLAER-Alexa488-based assays. The recent availability of a PE-conjugated form allowed us to test its utility in comparison with the predicate FLAER/CD24-based granulocyte assay.
CD157 is another GPI-linked structure, expressed on both granulocytes and monocytes, identified previously as a potential reagent for the detection of GPI-deficient cells ([25-27]). The recent availability of PE-conjugated CD157 allows direct comparison with CD24PE and with CD14PE in the 4-C granulocyte and 4-C monocyte protocols respectively.
In this study, we compare the predicate granulocyte assay with one in which CD66bPE was substituted for CD24PE and show herein that it is inferior to the predicate method. We have also compared the predicate granulocyte and monocyte assays with CD157-based equivalents on PNH and normal samples, and show that the CD157 assays perform in an equivalent manner with similar sensitivities on PNH samples and similar low background rates on normal samples.
For those with cytometers equipped with 5 or more photomultiplier tubes (PMTs), it is possible to use CD157 in combination with FLAER, CD15, CD64, and CD45 to detect both granulocytes and monocytes with high sensitivity in a single 5-color tube.
Finally, we have compared the utility of the predicate 4-color granulocyte and monocyte assays with the 4- and 5-color CD157-based assays on stabilized whole blood from PNH patients and show that the CD157 assays perform equally as well as the 4-C predicate assays on such samples.
- Top of page
- MATERIALS AND METHODS
- LITERATURE CITED
The development and validation of sensitive, standardized, flow cytometric methodologies to detect PNH was until recently, hampered by the rarity of this disease and technical difficulties in the accurate identification of PNH cells. This was addressed in 2010 by the publication of the ICCS Guidelines for the diagnosis and monitoring of PNH by flow cytometry (). The subsequent publication of the ‘Practical Guidelines’ describing specific reagent cocktails and detailed analytic strategies has further promoted the use of simple yet robust assays to detect PNH red and white blood cells that can be deployed on a range of flow cytometers with 4 or more PMTs (). These assays could reliably detect PNH granulocyte and monocyte clones at levels of 0.02% and 0.04%, respectively.
In this study we have addressed the utility of CD66bPE and CD157PE as potential alternative GPI-specific structures to CD24PE in the granulocyte assay and CD14PE in the monocyte assay. Unfortunately, we found the separation between PNH and normal granulocytes with CD66b to be sub-optimal compared with the predicate CD24PE reagents (clone ALB9, Beckman Coulter and clone SN3, eBioscience). Therefore we do not endorse the use of CD66b as an improved reagent over the currently recommended CD24.
CD157, a member of the CD38 supergene family and an ectoenzyme, functions as an integrin receptor and signaling molecule (). It is known to be expressed at relatively high antigen density on neutrophils and monocytes ([25-27]), although these studies were performed with a FITC conjugate of a different clone (RF3, Beckman Coulter). We show here that CD157-based assays perform at least as well as our CD24- and CD14-based predicates with PNH samples containing a wide range of clone sizes. Furthermore, background rates of PNH phenotypes in normal samples showed the same very low levels as the predicate assays. Overall, the CD157-based assays performed equally as well as the existing 4-color predicate assays previously published () that form the basis of recommendations on PNH testing in the Clinical and Laboratory Standards Institute H52-A2 document (). All 4-color assays described in this study also performed with similar efficiency on the FACSCalibur, platform either in combination with FLAER, CD15PECy5 (or CD64PECy5 for the monocyte assay) and CD45APC, or with FLAER, CD45PerCP, and CD15APC (or CD64APC for the monocyte assay) ().
The ability of CD157 to simultaneously serve as a GPI-specific marker for both neutrophil and monocyte lineages raised the possibility of a single tube 5-color PNH WBC assay. We found that the 5-color CD157-based assay detects both PNH neutrophils and PNH monocytes with high sensitivity and with the same low background rates on normal samples as noted with the predicate 4-color assays (). This is advantageous for laboratories with flow cytometers equipped with 5 or more PMTs, which perform frequent testing for PNH and related disorders. Indeed, the 5-color assay described herein, based on FLAER, CD157PE, CD64ECD, CD15PECy5, and CD45PECy7 generated equivalent data to the predicate and CD157-based 4-color assays on samples containing a wide range of PNH clone sizes on our FC500 instruments equipped with a single laser and 5 PMTs. This same combination was also assessed on a Navios cytometer equipped with 10 PMTs, deriving virtually identical data on the same samples.
In other experiments, the 5-color cocktail of FLAER, CD157PE, CD45PerCP, CD64APC, and CD15eFluor450 was capable of performing the same high-sensitivity analysis on a FACSCanto II instrument (data not shown).
One aspect in which the CD157-based assays differed from the CD24- and CD14-based predicates was in the ability of CD157-based assays to better delineate type II and type III granulocytes and type II and type III monocytes. While the clinical significance of the presence of type II white blood cells is unknown, it is important to include these cells in the overall assessment of PNH clone sizes for both granulocyte and monocyte lineages.
In our previous publication (), we recommended that laboratories use a tiered or sequential approach to the analysis of samples submitted for PNH testing. This approach involved testing all samples with the high sensitivity 2-color RBC and 4-color granulocyte assays, followed by a reflex monocyte assay in cases where PNH phenotypes were identified with RBC and/or granulocyte assays. In this setting, we recommended pre-cocktailed 2-color RBC and pre-cocktailed 4-color granulocyte reagents, while the reflex monocyte reagents were only utilized on an ‘as needed’ basis. We continue to believe that this is an efficient and cost-effective approach for laboratories that do not perform frequent PNH testing, as recently underscored by the study of Marionov et al. who demonstrated good intra- and inter-laboratory correlation, precision and reproducibility between laboratories in a multi-center study, independent of PNH clone sizes ().
However, for laboratories in major centers or reference laboratories that perform routine testing for PNH and related diseases, the single tube 5-color approach we describe herein has some significant advantages. Only a single 5-color WBC reagent cocktail is needed, thereby reducing sample preparation time, opportunities for pipetting errors, technical time, time needed for data acquisition/analysis, and significantly improving overall laboratory efficiency.
While not a critically important aspect of this study, stabilized whole blood samples are increasingly used by EQA schemes such as UK NEQAS ([21, 33]), QMP-LS (), and others. As shown above, all 4-color assays used in this study were able to detect and enumerate PNH granulocyte and monocyte phenotypes in a series of stabilized whole blood samples manufactured to contain a range (0.5%, 3% and 5%) of PNH granulocyte clones (22, MK and DRS unpublished data). Similarly, the CD157-based 5-color assay was able to accurately enumerate PNH granulocytes and monocytes in these stabilized samples. We previously assessed the stabilized samples with the high-sensitivity RBC assay and found this material to be suitable for the assessment of PNH RBCs (). Thus, all WBC and RBC assays described herein and previously () can be used with stabilized PNH whole blood preparations that are increasingly used in EQA programs.