|
 
 |
| ORIGINAL ARTICLE |
|
| Year : 2018 | Volume
: 9
| Issue : 3 | Page : 85-90 |
|
Clinical and flow cytometric analysis of paroxysmal nocturnal hemoglobinuria in Indian patients
Rajesh Kashyap1, Namrata Punit Awasthi2, Ritu Gupta3
1 Department of Hematology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, Uttar Pradesh, India
2 Department of Hematology, Sanjay Gandhi Postgraduate Institute of Medical Sciences; Department of Pathology, Dr. Ram Manohar Lohia Institute of Medical Sciences, Lucknow, Uttar Pradesh, India
3 Department of Hematology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, Uttar Pradesh; Department of Laboratory Oncology, BIRCH, All India Institute of Medical Sciences, New Delhi, India
| Date of Web Publication | 31-Oct-2018 |
Correspondence Address:
Dr. Rajesh Kashyap
Department of Hematology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow - 226 014, Uttar Pradesh
India
 Source of Support: None, Conflict of Interest: None  | Check |
DOI: 10.4103/joah.joah_11_18
INTRODUCTION: Paroxysmal nocturnal hemoglobinuria (PNH) is an uncommon disease. Many cases go undiagnosed as high index of clinical suspicion is required for its detection. This study was performed to detect the presence of PNH defect by flow cytometric immunophenotyping (FCMI) in patients with suspected PNH disease and evaluate their clinical and laboratory profile.
MATERIALS AND METHODS: In this retrospective study, a total of 136 patients with suspected PNH who fulfilled the inclusion criteria for the study were evaluated for PNH defect by FCMI using monoclonal antibodies against CD55 and CD59 on red blood cell, granulocytes, and monocytes.
RESULTS: Forty-eight (35.3%) of 136 patients evaluated had a PNH defect. Nineteen (39.5%) of these 48 patients had classical PNH (hemolytic type). The remaining 29 patients had PNH Clone in association with aplastic anemia. The clinical and laboratory data of these 19 patients with classical PNH were analyzed in this retrospective study. The median age was 34 years (range: 19–65 years). Thrombotic events were observed in 3 (16%) of the 19 cases (one each with Budd–Chiari syndrome, cerebral venous thrombosis, and abdominal vein thrombosis). The flow cytometric data of these patients were further analyzed for the presence of and size of PNH clone on erythrocytes, granulocytes, and monocytes. PNH clone was detected in 84% of erythrocytes, 76.9 % of monocytes and in 100% granulocytes.
CONCLUSION: Classical PNH is not rare in India as previously thought. A high index of clinical suspicions and evaluation by FCMI is necessary for its detection. CD59 is a better marker for identification of PNH clone than CD55 in all three cell types. Keywords: CD55, CD59, flow cytometric immunophenotyping, glycosylphosphatidylinositol-anchored proteins, hemolysis
How to cite this article:
Kashyap R, Awasthi NP, Gupta R. Clinical and flow cytometric analysis of paroxysmal nocturnal hemoglobinuria in Indian patients. J Appl Hematol 2018;9:85-90 |
How to cite this URL:
Kashyap R, Awasthi NP, Gupta R. Clinical and flow cytometric analysis of paroxysmal nocturnal hemoglobinuria in Indian patients. J Appl Hematol [serial online] 2018 [cited 2023 Mar 24];9:85-90. Available from: https://www.jahjournal.org/text.asp?2018/9/3/85/244533 |
| Introduction | |  |
Paroxysmal nocturnal hemoglobinuria (PNH) is a clonal disorder caused by somatic mutation of the X-linked phosphatidylinositol glycan-class A (PIG-A) gene resulting in deficient expression of glycosylphosphatidylinositol-anchored proteins (GPI-APs).[1] Deficiency of GPI-linked complement regulatory proteins, i.e., membrane inhibitor of reactive lysis (CD59), decay-accelerating factor (CD55), and other proteins is responsible for the marked susceptibility of red blood cells (RBCs) of PNH to the complement-mediated hemolysis leading to the characteristic clinical feature of intravascular hemolytic anemia.[2] PNH is largely underdiagnosed because of varied clinical presentation and atypical symptoms. The International PNH Interest Group (I-PIG) has proposed a classification system for PNH as follows:[3] (1) classical PNH – patients having clinical and laboratory features of overt intravascular hemolysis, a cellular bone marrow with erythroid hyperplasia, and large PNH clones; (2) PNH in setting of another specific bone marrow disorder (aplastic anemia or myelodysplastic syndrome) – patients having clinical and laboratory evidence of hemolysis in a setting of specified bone marrow disorder; and (3) subclinical PNH (PNH-sc) – no evidence of hemolysis but small PNH clones detected by flow cytometry (typically seen in aplastic anemia).
Flow cytometric immunophenotyping (FCMI) is now widely used for the diagnosis of PNH as it can detect the presence of PNH clone in all the hematopoietic cell lineages, i.e., erythrocytes, leukocytes, and platelets and also provides an estimate of PNH subpopulations in erythrocytes, i.e., PNH Type I, II, and III cells depending on the intensity of GPI-APs on cell surface. The PNH Type I erythrocytes have normal levels of GPI-APs on their cell surface; PNH Type II have intermediate expression; and PNH Type III cells have complete deficiency of GPI-APs.[4]
Reports of classical PNH from Indian subcontinent consist of isolated case reports and few retrospective analyses of clinical and laboratory findings.[5],[6],[7],[8] The present study was undertaken to study the clinical behavior, flow cytometric findings, and outcome of classical PNH in Indian patients.
| Materials and Methods | | |
This retrospective study was conducted in the Department of Hematology, Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow, India. A total of 136 patients over a period of 6 years from June 2009 to May 2015 who had any one or more of the following clinical and laboratory features were evaluated and screened for PNH of (1) macrocytic anemia refractory to treatment with parenteral Vitamin B12 and folic acid therapy for 4–8 weeks' duration, (2) bicytopenia/pancytopenia with or without hepatosplenomegaly and no lymphadenopathy, (3) hemolysis (anemia with or without jaundice, dark-colored urine) with reticulocytopenia and negative direct antiglobulin (Coombs test) test, and (4) systemic or cerebral venous or arterial thrombosis.
The clinical data, treatment, and follow-up data of all the patients were retrieved from their medical records of all the patients. The laboratory investigations included complete blood counts, reticulocyte count, red cell indices, serum bilirubin, serum lactate dehydrogenase (LDH), direct and indirect Coombs test, urine hemosiderin, plasma hemoglobin, bone marrow examination, and FCMI.
Flow cytometric analysis
The PNH clone was evaluated both in erythrocytes as well as leukocytes using FCMI. Two milliliters of ethylenediaminetetraacetic acid-anticoagulated peripheral blood was collected from each individual, and sample was processed and acquired within 6 h of sampling. In case of delay in processing, samples were stored at 4°C. For erythrocytes, a single-wash-no-lyse technique was used. Briefly, 10-μL whole blood was diluted in 1 ml of phosphate-buffered saline (PBS). About 100 μL of this diluted RBCs was taken and incubated with pretitrated volumes of phycoerythrin (PE)-conjugated monoclonal antibodies, i.e., CD55 (IA10, Pharmingen, BD Biosciences, San Diego, CA, USA) and CD59 (H 19, Pharmingen, BD Biosciences) in the dark at room temperature for 20 min, followed by washed with PBS, resuspended in 1% paraformaldehyde. For leukocytes, a stain-lyse-wash protocol was used for all the cases. Lysis of RBCs was carried out with an ammonium chloride before incubation with CD16 (fluorescein isothiocyanate), CD55 (PE), and CD59 (PE).
The acquisition was performed on FACSCanto II system and analysis was performed using FACSDiva software (BD Biosciences, USA). A minimum of 5000 RBCs and 5000 white blood cells (WBCs) were acquired. In case of leukopenia, it was attempted to acquire at least 2000 neutrophils and a maximum number of monocytes as possible.
Cell populations were labeled as PNH clone on FCMI when there was evidence of a population of peripheral blood cells (erythrocytes, granulocytes, or preferably both) deficient in GPI-APs.[3] The cutoff positivity for PNH was taken as 1%. Based on the mean relative intensity of CD55 or CD59 fluorescence, erythrocytes were separated into three discrete populations, i.e., PNH Type I cells as those with an intensity that lays within the normal range (defined as mean ± 2 standard deviation [SD]) calculated from normal healthy controls (n = 20); PNH Type III as cells with intensity within the CD55- or CD59-negative area in the flow cytometric profile, defined as the area containing 99.8% of the negative isotype control cells; and PNH Type II cells as those with intensity of CD55 and/or CD59 in-between Type I and III cells [Figure 1]. The granulocytes and monocytes were identified by their characteristic light scatter properties on dot plot scattergrams and evaluated for the expression of CD55 and CD59 with reference to the normal control samples processed on the same day or remaining normal granulocytes and monocytes within the same sample [Figure 2]. | Figure 1: Flow cytometric histograms showing expression of CD55 and CD59 on erythrocytes of healthy participant (a and b) and in a patient with paroxysmal nocturnal hemoglobinuria (c and d). M1, M2, and M3 denote Type I, Type II, and Type III paroxysmal nocturnal hemoglobinuria cells, respectively
Click here to view |
 | Figure 2: Flow cytometric dot plots showing gating of neutrophils (a) based on forward scatter and side scatter followed by expression of CD55, CD59, and CD16 on granulocytes of a patient with paroxysmal nocturnal hemoglobinuria (b and c)
Click here to view |
Statistical analysis
Descriptive statistical analysis of main characteristics of patients was performed. Categorical and continuous data were presented in frequency (%) and mean ± SD.
In this retrospective study, as there were no patient-identifying data and no new tests or interventions were performed, as per the institute's IRB rules and guidelines, no approval was required.
| Results | | |
During the study period, a total of 136 patients were evaluated for the presence of PNH clone. Forty-eight (35.3%) patients were detected to have PNH clone either in erythrocytes or leukocytes or both. Twenty-nine (60.4%) of these 48 patients had small PNH clone in the setting of aplastic anemia (PNH-sc) and were excluded from further analysis. The remaining 19 patients (14 males and 5 females) had features of classical PNH. The median age was 34 years (range: 19–65 years). Thrombotic events were observed in three (16%) of the 19 cases (one each with Budd–Chiari syndrome, cerebral venous thrombosis, and abdominal vein thrombosis) [Table 1]. Eleven patients (58%) had a history of one or more episodes of hemoglobinuria. In 11 patients, the diagnosis of PNH could be established soon after the initial presentation, whereas in 8 patients, the average delay between presentation and diagnosis was 3.8 years. | Table 1: Clinical and laboratory profile of patients with classical paroxysmal nocturnal hemoglobinuria
Click here to view |
Severe anemia (hemoglobin [Hb] level <8 g/dl) was found in 14 (74%) patients. The mean Hb concentration was 5.9 g/dl (range: 2.9–13.9 g/dl). Seven patients had a total leukocyte count (TLC) of less than 4 × 109/L and mean TLC of 5.61 × 109/L (range: 1–26 × 109/L). The mean platelet count in these patients was 172.57 × 109/L (range: 31–618 × 109/L). The mean value of corrected reticulocyte count was 3.4% ± 4.3%. Serum LDH was high in all patients with a mean value of 2488.58 IU/L (range: 133–9416 IU/L) [Table 1]. Urine hemosiderin was positive in four patients. Plasma Hb levels were raised in eight cases who had ongoing episode of hemolysis at the time of diagnostic workup. Bone marrow showed cellular marrow with erythroid hyperplasia in all the cases.
Flow cytometric analysis
Flow cytometric immunophenotypic analysis of erythrocytes, granulocytes, and monocytes was done using CD55 and CD59 [Table 2]. Follow-up FCMI analysis after 5 years was available in one case. | Table 2: Immunophenotypic analysis of erythrocytes, granulocytes, and monocytes for paroxysmal nocturnal hemoglobinuria clone size using CD55 and CD59 in patients with classical paroxysmal nocturnal hemoglobinuria (n=19)
Click here to view |
Flow cytometric analysis of erythrocytes
Of the 19 patients evaluated, PNH clone in erythrocytes was detected in 16 (84%) patients and PNH clone could not be identified in 3 patients [Table 2]: case no. 2, 7, and 8]. CD55-deficient erythrocytes were seen in 15 patients with Type III cells (ranging from 5% to 1%) and Type II cells (ranging from 2% to 45%). Mean clone size of CD55-deficient erythrocytes was 44.1% ± 19.9%. CD59-deficient erythrocytes were seen in 16 patients with Type III cells (ranging from 2% to 86%) and Type II cells (ranging from 1% to 48%). Mean clone size of CD59-deficient erythrocytes was 39.8% ± 26.6%. Isolated CD55 deficiency was not found in any patient, and isolated CD59 deficiency was found in one patient with 5% Type II PNH cells. Both granulocytes and monocytes revealed PNH clones in this patient [Table 2].
Flow cytometric analysis of granulocytes and monocytes
Out of the 19 patients, PNH clone in granulocytes was identified in all 19 patients. Mean clone size of CD55-deficient granulocytes was 50.2% (range: 4%–99%) and of CD59-deficient granulocytes was 63.3% (range: 1%–100%). Three patients of CD55-deficient granulocytes were identified in whom a corresponding PNH clone of CD59-deficient granulocytes was not identified (cases 7, 17, and 18), and four patients of CD59-deficient granulocytes were identified with no evidence of CD55-deficient PNH clone (cases 4, 5, 15, and 19) [Table 2].
PNH clone in monocytes was found in 15 out of 19 cases (78.9%). In four cases PNH clone could not be assessed because of very low population of monocytes in these samples. Mean clone size of CD55-deficient monocytes was 63% (range: 13%–100%) and of CD59-deficient monocytes was 72.2% (range: 43%–100%). One patient of CD55-deficient monocytes was identified in whom a corresponding PNH clone of CD59-deficient monocytes was not identified (case 7). Similarly, one patient of CD59-deficient monocytes was identified in whom a corresponding PNH clone of CD55-deficient monocytes was not identified (case 14).
All the cases showed variable numbers of Type I, Type II, and Type III erythrocytes. There were three patients in whom a PNH clone was not identified in erythrocytes but was identified in leukocytes (cases 2, 7, and 8). A comparative analysis of the size of PNH clone in all three cell types was done. It was seen that the mean clone size was largest in monocytes followed by granulocytes and then erythrocytes.
Follow-up
Follow-up data were available in 14 patients. Median duration of follow-up was 36.6 months. One patient with cerebral venous thrombosis expired after 4 months. Twelve out of 13 patients were treated with prednisolone and/or danazol and showed satisfactory response for rise in Hb; however, the response for rise in TLC and platelet count was seen in only 2 patients.
| Discussion | | |
Although PNH is a well-recognized hematological disorder, it is still largely underdiagnosed due to its variable clinical presentation. After the availability of more sensitive tests such as FCMI for quantitation and identification of small PNH clone, screening and diagnosis of patients with PNH has considerably improved. The annual incidence of PNH in the Western countries is reported to be 0.13 per 100,000. PNH appears to be more frequent in some Asian countries such as China and Thailand, than in Western countries.[9] The prevalence of disease in the Indian population is not known, although there are few Indian studies, indicating that the disease is not rare in Indians.[5],[6],[7],[8] In the present study, the incidence of classical PNH was 39.5%. However, the incidence of classical PNH was 11.8% in a study by Rahman et al.[10] The difference in the incidence is probably due to the use of more sensitive diagnostic techniques used in later study.
We diagnosed 19 cases of classical PNH in our hospital over duration of 6 years. The mean age of our patients was 36.05 years which is slightly higher than that reported by Koduri et al. and Gupta et al.[7],[8] but in agreement with the findings of Zhao et al.[11]. There was male predominance in our study with M:F ratio of 2.8:1, and similar male predominance has also been reported in other Indian,[5],[6],[7],[8],[10],[12] Thai,[9] and Chinese[11] studies. Epidemiological studies in Europe and the United States suggest that the incidence of thrombosis is approximately 40%,[13] whereas the incidence in East Asian countries is considerably less, namely 3%–6%.[11] The incidence of thrombosis in our study was 16%. The basis of this phenotypic difference is unknown, but the relationship of ethnicity and geography to the natural history of PNH should be considered when formulating a management plan.[8]
Hemoglobinuria was found in 58% of our patients. It is the presenting symptom in majority of adult patients.[7],[8],[11] Severe anemia (Hb <8 g/dl) was the predominant finding in peripheral blood seen in 74% of our cases, and the mean reticulocyte count was 3.4% ± 4.3%. This indicates a blunted reticulocyte response in relation to the degree of anemia which is in all probability due to some degree of underlying bone marrow failure. This finding is in sync with the fact that aplastic anemia and PNH are two closely related disorders and one may progress to another in due course of time. Many patients with PNH develop aplastic anemia as the final stage of disease,[13] and PNH eventually develops in 10%–31% of patients with aplastic anemia treated with immunosuppressive therapy.[14] Majority of patients with PNH probably have an underlying aplastic process.[13],[15] As we had excluded cases of aplastic anemia from our study, the most common finding on bone marrow examination was erythroid hyperplasia with no case showing hypocellular marrow.
Flow cytometry continues to play a pivotal role in diagnosis and management of PNH. In our study, 91% of our patients showed Type III. The similar observation was noted by Pramoonjago et al.[9] in their study. However, our data showed more number of cases with Type II PNH clone on erythrocytes, and this finding is supported by Hill et al. who have described that at least 40% patients have a combination of PNH Type I, II, and III cells.[16] Another interesting finding in our study is that the mean clone size in monocytes was found to be more than in granulocytes for both CD55 and CD59; however, the difference is not statistically significant which in all probability is due to small sample size. Further studies directed toward this difference in clone size and its impact on PNH pathophysiology need to be carried out.
In the present study, mean clone size of erythrocytes for both CD55 and CD59 was low as compared to mean clone size for granulocytes and monocytes. This difference is due to ongoing hemolysis of PNH erythrocytes and a variable proportion of transfused normal red cells. Moreover, phagocytes (granulocytes and monocytes) are relatively durable during laboratory processing, whereas erythrocytes are known to hemolyse easily. The optimal approach, therefore, appears to be simultaneous evaluation of erythrocytes and granulocytes[16],[17] as well as monocytes for the expression of CD55 and CD59. CD59 comes out to be a better marker for identification of PNH clone in all three cell types in our study as also suggested by other researchers.[18] This finding was further confirmed in three of our patients in whom a PNH clone was not identified in erythrocytes but was identified in leukocytes (cases 2, 7, and 8). Several studies have suggested that size of PNH clone is a factor which determines severity of disease, but probably, there are other factors also such as hematopoietic activity and environmental factors such as infections which exacerbate hemolysis.[9]
Guidelines for diagnosis and monitoring of PNH disorders by flow cytometry have been published by Borowitz et al.[19] which was followed by practical guidelines to perform high-sensitivity assay by Sutherland et al.[20] Sensitivity for detection of PNH clone for routine analysis has been suggested as 1% which was achieved in the present study. High-sensitivity assays are not needed for diagnosis of classic PNH. In the current study, gating antibodies were not utilized for identifying erythrocytes, granulocytes, and monocytes, and these populations were identified based on the scatter properties. This is a limitation of the current study as addition of gating antibodies results in better distinction of cell populations. Another limitation is that Fluorescently labelled aerolysin assay (FLAER) was not used in this study which is perhaps the most useful reagent for detecting PNH clones in WBC.
| Conclusion | | |
PNH is a disease with varied clinical manifestation and is not rare in India as previously thought. A high index of clinical suspicions and evaluation of erythrocytes, granulocytes, and monocyte by FCMI is necessary for its detection. CD59 is a better marker for identification of PNH clone than CD55. The role of the size of PNH clone of these three cell types and its impact on PNH pathophysiology needs to be studied on a larger sample size.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
| References | | |
| 1. | Takeda J, Miyata T, Kawagoe K, Iida Y, Endo Y, Fujita T, et al. Deficiency of the GPI anchor caused by a somatic mutation of the PIG-A gene in paroxysmal nocturnal hemoglobinuria. Cell 1993;73:703-11.  |
| 2. | Rosse WF. Paroxysmal nocturnal hemoglobinuria as a molecular disease. Medicine (Baltimore) 1997;76:63-93. |
| 3. | Parker C, Omine M, Richards S, Nishimura J, Bessler M, Ware R, et al. Diagnosis and management of paroxysmal nocturnal hemoglobinuria. Blood 2005;106:3699-709. |
| 4. | Fujioka S, Yamada T. Varying populations of CD59-negative, partly positive, and normally positive blood cells in different cell lineages in peripheral blood of paroxysmal nocturnal hemoglobinuria patients. Am J Hematol 1994;45:122-7. |
| 5. | Parab RB. Paroxysmal nocturnal hemoglobinuria: A study of 17 cases. J Postgrad Med 1990;36:23-6. [ PUBMED] [Full text] |
| 6. | Agarwal MB, Mehta BC. Paroxysmal nocturnal hemoglobinuria: (A report of 20 cases). J Postgrad Med 1981;27:231-4. [ PUBMED] [Full text] |
| 7. | Koduri PR, Gowrishankar S. Paroxysmal nocturnal haemoglobinuria in Indians. Acta Haematol 1992;88:126-8. |
| 8. | Gupta PK, Charan VD, Kumar H. PNH revisited: Clinical profile, laboratory diagnosis and follow-up. Indian J Pathol Microbiol 2009;52:38-41. [ PUBMED] [Full text] |
| 9. | Pramoonjago P, Pakdeesuwan K, Siripanyaphinyo U, Chinprasertsuk S, Kinoshita T, Wanachiwanawin W, et al. Genotypic, immunophenotypic and clinical features of thai patients with paroxysmal nocturnal haemoglobinuria. Br J Haematol 1999;105:497-504. |
| 10. | Rahman K, Gupta R, Yadav G, Husein N, Singh MK, Nityanand S, et al. Fluorescent aerolysin (FLAER)-based paroxysmal nocturnal hemoglobinuria (PNH) screening: A single center experience from India. Int J Lab Hematol 2017;39:261-71. |
| 11. | Zhao M, Shao Z, Li K, Chen G, Liu H, Zhang Y, et al. Clinical analysis of 78 cases of paroxysmal nocturnal hemoglobinuria diagnosed in the past ten years. Chin Med J (Engl) 2002;115:398-401. |
| 12. | Sachdeva MU, Varma N, Chandra D, Bose P, Malhotra P, Varma S, et al. Multiparameter FLAER-based flow cytometry for screening of paroxysmal nocturnal hemoglobinuria enhances detection rates in patients with aplastic anemia. Ann Hematol 2015;94:721-8. |
| 13. | Hillmen P, Lewis SM, Bessler M, Luzzatto L, Dacie JV. Natural history of paroxysmal nocturnal hemoglobinuria. N Engl J Med 1995;333:1253-8. |
| 14. | Tichelli A, Gratwohl A, Würsch A, Nissen C, Speck B. Late haematological complications in severe aplastic anaemia. Br J Haematol 1988;69:413-8. |
| 15. | Rosse WF, Nishimura J. Clinical manifestations of paroxysmal nocturnal hemoglobinuria: Present state and future problems. Int J Hematol 2003;77:113-20. |
| 16. | Hill A, Richards SJ, Hillmen P. Recent developments in the understanding and management of paroxysmal nocturnal haemoglobinuria. Br J Haematol 2007;137:181-92. |
| 17. | Oelschlaegel U, Besson I, Arnoulet C, Sainty D, Nowak R, Naumann R, et al. A standardized flow cytometric method for screening paroxysmal nocturnal haemoglobinuria (PNH) measuring CD55 and CD59 expression on erythrocytes and granulocytes. Clin Lab Haematol 2001;23:81-90. |
| 18. | Tembhare P, Ramani M, Syed K, Gupta AD. Flow cytometric analysis of erythrocytes in paroxysmal nocturnal hemoglobinuria reveals superiority of CD59 as a diagnostic marker compared to CD55. Indian J Pathol Microbiol 2010;53:699-703. [ PUBMED] [Full text] |
| 19. | Borowitz MJ, Craig FE, Digiuseppe JA, Illingworth AJ, Rosse W, Sutherland DR, et al. Guidelines for the diagnosis and monitoring of paroxysmal nocturnal hemoglobinuria and related disorders by flow cytometry. Cytometry B Clin Cytom 2010;78:211-30. |
| 20. | Sutherland DR, Keeney M, Illingworth A. Practical guidelines for the high-sensitivity detection and monitoring of paroxysmal nocturnal hemoglobinuria clones by flow cytometry. Cytometry B Clin Cytom 2012;82:195-208. |
[Figure 1], [Figure 2]
[Table 1], [Table 2]
|
Search Pubmed for