Background: Blood cells are deficient in membrane-bound glycosyl-phosphatidylinositol (GPI)-anchored proteins in paroxysmal nocturnal hemoglobinuria (PNH) because of an acquired disorder of the pluripotent stem cell. CD55 and CD59 antigens are the most common GPI-anchored proteins that are used for the diagnosis of the presence of PNH clone. The association of MDS with PNH is rather controversial. There are few published individual cases of PNH arrived from a previous MDS or MDS following PNH.

Aim: Recent studies have demonstrated the existence of PNH clone in some MDS patients. These studies were performed mostly on erythrocytes and to the best of our knowledge they are only 4. Granulocytes appear to be more sensitive markers of PNH clone existence than erythrocytes and therefore we investigate the expression of PNH clone on granulocytes of patients with MDS.

Material-Methods: A total number of 95 patients 25–80 yrs old with MDS {A:19 with refractory anaemia (RA), B:9 with refractory anaemia with ring sideroblasts (RARS), C:17 with refractory anaemia with excess blasts (RAEB), D: 12 with refractory anaemia with excess blasts in transformation (RAEB-t) and E: 19 with chronic myelomonocytic leukemia (CMLL)} and 19 healthy donors were included in our study. The presence of GPI-anchored proteins (CD55, CD59)-deficient granulocytes was examined by flow cytometry. For the detection of the PNH clone the commercial kit by Beckman Coulter (cellquant CD55/CD59) was used. Statistical analysis was made by ANOVA, while Robust test was performed because there was no homogeneity of variances by ANOVA.

Results: Table 1 shows the percentages of CD55 and CD59 deficient granulocytes in the 5 groups of pts with MDS comparing with the control group. a) Although the mean values of CD55 deficient granulocytes in the groups RAEB, RAEB-t and CMLL were higher enough than those of the normal subjects, there were no statistically significant differences (p>0.10) in the comparison of the mean values of CD55 deficient granulocytes between each group of patients with the control group. b) On the contrary there was a very statistically significant difference in the comparison of the mean value of CD59 deficient granulocytes between the three last groups (C, D and E) of pts with the control group (p< 0.086, p< 0.001 and p<0.001 respectively).

The percentages of CD55 and CD59 deficient granulocytes in the 5 groups of patients with MDS and in the control group

CD55CD55CD59CD59
Groups Mean Value Std deviation Mean Value Std deviation 
A (RA) 19 2.89 2.79 2.91 2.85 
B (RARS) 1.05 1.14 3.89 3.63 
C (RAEB) 17 4.78 4.76 11.75 6.06 
D (RAEB-t) 12 4.23 2.85 23.76 8.53 
E (CMLL) 19 6.01 4.97 38.26 16.18 
Control 19 3.67 2.89 3.67 2.89 
Total 95 4.00 3.87 14.44 16.39 
CD55CD55CD59CD59
Groups Mean Value Std deviation Mean Value Std deviation 
A (RA) 19 2.89 2.79 2.91 2.85 
B (RARS) 1.05 1.14 3.89 3.63 
C (RAEB) 17 4.78 4.76 11.75 6.06 
D (RAEB-t) 12 4.23 2.85 23.76 8.53 
E (CMLL) 19 6.01 4.97 38.26 16.18 
Control 19 3.67 2.89 3.67 2.89 
Total 95 4.00 3.87 14.44 16.39 

Conclusions: There are relatively large CD59 negative subpopulations (mean values 11.75–38.26%) of granulocytes in the last three groups (RAEB, RAEB-t, CMLL) of MDS patients. On the contrary the two low risk forms (RA, RARS) revealed no such populations. Non-expression of CD59 antigen on granulocytes is a more sensitive marker than that of CD55 antigen for the presence of a possible PNH clone in MDS patients.

Author notes

Corresponding author

Sign in via your Institution