As a result of many requests, I am posting the numerous replies I receive to my request for information on diagnosing PNH by flow. I have not included names and numbers of people who were suggested for me to contact. I received a lot of information including markers to try as well as references. To summarize, the expression of CD59 is always examined on at least red cells and usually granulocytes as well. Other labs include CD55, CD14, CD66b, CD24, etc. Thanks again to all of those ( and there were many!) who replied for your very informative and timely advice! Sincerely, Abby Kelliher Flow Cytometry Lab - Mass General Hospital RESPONSES: This not a simple test, but it can be done...we do it routinely here at Penn..you need to run a normal along side the patient and analyze the RBC's, lymphs, grans and monos for CD55 and 59. on histograms. PNH is a clinical diagnosis so what you can detect are PNH clones which are consistent with the disease but not diagnostic. Remember, you are looking for DIMINISHED EXPRESSION OF THESE MARKERS indicating loss of GPI linked protein. If you have not done this before, I would suggest not to start with a patient since you have to qc your reagent, determine regions, etc. When we evaluate PB specimens for PNH we look for the lack of expression of CD59 on red cells in conjunction with a conjugated anti-Glycophorin-A. We also look for the lack of both CD55 and CD59 on CD13 positive neutrophils. So far I have only seen one positive case, needless to say it was quite exciting. As a reference try Blood, Vol 87, No. 12 (June 15) 1996: pp 5332-5340 Titled "The Use of Monoclonal Antibodies and Flow Cytometry in the Diagnosis of Paroxysmal Nocturnal Hemoglobinuria" by Hall and Rosse. PNH is a disease that lacks the ability to express phosphatidyl-linked membrane proteins (ie.DAF, HRF, CD59). Intravascular hemolysis partially due to completment activation. You want to look for the absence of PI-linked antigens on the PNH cells some of which are CD59, CD14, CD66b to name a few..... CD55 and CD59 on red cells I would try using CD59 measure GPI-linked proteins (with CD59FITC) on red cells and granulocytes by flow cytometry. It provides a specific and sensitive technique for screening and diagnosis of PNH. most commonly applied: CD59 next most commonly applied: CD59 and CD55 also possible: many other GPI-linked proteins are commonly found among reagents in a flow lab - CD14, CD24, CD16(on PMNs, not the CD16 on NKs, they are transmembrane proteins). Most typical cell to test is the RBc. However, if there has been a lysis crisis, particularly followed by transfusion of normal RBCs, PMNs might be a good cell to also test. Perform CD55 and CD59; must be performed on fresh specimens The lab I used to work in routinely used CD59, CD55, CD16, and CD14. You need to not only look at RBCs but should also investigate WBCs as the deficiency can be found in any of the lineages (although lymphs are rare). If you only investigate RBCs you might miss it. See the response below from the cytometry archives: I would like to thank everyone who answered my question about the involvement of different cell lineages in PNH. I will try to briefly summarize the information: "The consensus seems to be that loss of PI linked markers is most often seen in RBC and granulocytes, then monocytes, and least often in lymphocytes. Because lymphocytes are seldom affected some feel it is not useful to examine them routinely. There is no clear information about platelets. It may be possible to pick up the defect in marrow cells before it is manifested in the blood. Some respondents pointed out that the defect can be manifested as an incomplete loss of a marker, i.e. the cells may show decreased mean fluorescence on the entire population, or there may be a subset with decreased or absent expression, or there may be absent expression on the entire population. Among the PI linked markers that have been used are: CD59, CD48, CD58, CD67 (now CD66b), CD55, CD16, and CD14. It is important to test several of them because they may not all be equally affected, and not all cell lineages are equally affected. Finally, the expression of the defect can change over time, with different lineages becoming affected. There is a report of a few patients whose erythrocytes reverted to normal, but the lymphocytes continued to be abnormal. References: Nakakuma et al Blood 85:1371-1376, 1995. Hillmen et al NEJM 333:1253-1258, 1995. Griscelli-Bennaceur et al Blood 85:1354-1363, 1995. Hall and Rosse Blood 87:5332-5334, 1996. BeckmanCoulter offers CD55/CD59 determination kits which are helpful with PNH. You can test for the presence of CD59 or 54 on red cells, CD66b and CD16 on granulocytes or CD14 on monocytes. According to the literature, CD66b on granulocytes is the most sensitive - possibly to < 2%, but you need to do a 2-3 color combination with CD15 to identify granulocytes along with side scatter, and CD66b and CD16. When you plot CD66b vs CD16, gating on CD15 vs SS, if there are more than 2% of cells in the negative quadrant #3, it is PNH. I would run a few normal controls along with the patient. Loss of CD16 alone is not important, since the granulocytes shed this, but will still express CD66b. CYTOMETRY Vol 38, #6, 1999 pp259-267 or, www3.interscience.wiley.com/cgi-bin/abstract?IDh500892 bunch of refs. CD14 on monocytes, CD16 on granulocytes, and CD55 and CD59 on red cells are all depressed in PNH. These markers are readily detected by flow cytometry. Mayo Clinic Labs offers a package for testing if these markers are not available in your own flow lab. we only use CD55 and 59 as marker and we only look at the neutrophils and erythrocytes. some protocol also use CD14 to include monocytes, CD13, CD16, CD45 can also be included. Check ref. The use of Monoclonal Antibodies and Flow Cytometry in the Diagnosis of PNH by Sharon E.Hall and Wendell F. Rosse. CD55 and CD59 are the GPI-linked complement regulatory molecules deficient in PNH patients. Hence the hemolysis, I don't recall why it happens at night. Beckman Coulter has a kit designed to give quantitative measurement of both CD55 and CD59 molecules on either RBC (Part# BX7201) or on granulocytes (Part# BX73901).
This archive was generated by hypermail 2b29 : Sat Mar 10 2001 - 19:31:03 EST