Case 362 -- Myelodysplastic syndrome and bone marrow transplant engraftment failure

Contributed by Mark K Fung, MD, PhD
Published on line in September 2003


CLINICAL HISTORY:

Patient was a man in his 60s who developed myelodysplastic syndrome after chemotherapy (MOP) and autologous bone marrow transplant for Hodgkins disease. The patient therefore needed an allogeneic bone marrow transplant as part of the next phase of his treatment.

No HLA-identical siblings were identified, therefore a search for an unrelated marrow allograft donor was initiated through the National Marrow Donor Program (NMDP). A donor was identified and stem cells were collected for MUD-BMT.

The patient's and the NMDP donor's HLA typings are listed below:

 

A*

B*

Bw

C*

DRB1*

other DR

DQB1*

Patient

0201

32BMP

27XX

44XX

4

-

0102

05RV

0102

1501

5*XX

-

0501

06XX

NMDP

Donor

32XX

33DWH

27XX

44FGG

-

-

0102

0202

0102

1501

5*XX

-

0501

06XX

*Age of patient and HLA typings changed to protect patient's privacy.
HLA-Bw typing is based on common associations with HLA-A and -B antigen types

The patient's quick cytotoxic panel reactive antibody (PRA) screen was negative for anti-HLA antibodies. (PRA = Panel reactive antibodies = panel of different T-lymphocytes with representative HLA class I antigens plated on a 60 well plate mixed with patient serum. Lymphocytes that become bound with antibodies are lysed in the presence of complement that is also added. Reactive wells are scored on the basis of the percentage of dead cells identified by a viability stain).

Both the B- and T-cell crossmatches (donor lymphocytes mixed with patient serum) were negative by complement-mediated lymphocytotoxic assays. The T-cell crossmatch is performed with anti-human immunoglobulin antibody (AHG) enhancement to increase the sensitivity of detecting anti-HLA antibodies against the donor lymphocytes. No AHG enhancement was used with the B-cell crossmatch. Given these favorable results, the MUD-BMT procedure was undertaken using stem cells collected from the above NMDP donor.

However, one month after MUD-BMT, the patient was experiencing engraftement failure. At this point his quick PRA showed 100% reactivity against lymphocytes of various HLA types using serum treated with and without dithiothreitol (DTT, used to break the disulfide bonds of IgM-pentameric antibodies. IgG antibodies are more resistant to DTT treatment). In addition, the T-cell crossmatch with AHG enhancement showed strong positive reaction at 1:1 titer of patient serum with weaker reactions at higher dilutions of serum. The B-cell crossmatch was negative at all concentrations of patient serum.

Question: Is this consistent with an immune-mediated rejection against the mismatched donor antigens?

Answer: Yes, the mismatched antigens: A*33DWH and C*0202 are class I antigens. There were no mismatched class II antigens. Class I antigens (HLA-A, B, C) are found on all cells and T-lymphocytes. Class II antigens (HLA-DR, DQ, DP) are found only on B-lymphocytes and certain activated immune cells.

Advanced Question: If there are antibodies against HLA class I antigens, should not B-cells also be reactive in a crossmatch?

Answer: Yes! B-cells should be reactive because they also contain class I antigens. In fact, B lymphocytes contain a higher density of HLA class I antigens than do T lymphocytes.

Advanced Question: Why then did the B-cells not react to the presence of antibodies against HLA class I antigens?

Answer: There are some possible explanations. First, the antibody titer was weak (positive only to a 1:1 dilution) and was detectable only with AHG enhancement in the T-cell crossmatch. Second, the B-cell crossmatch was performed in this instance without enhancement with anti-human antiglobulin. A third possibility is further discussed later in the case.

Question: How does anti-human antiglobulin enhance the lymphocytotoxic reaction performed in the T-cell crossmatch and in the quick PRA?

Answer: The lymphocytotoxic reaction used in crossmatches and in the quick PRA depend on complement activation of antibodies bound to the surface of lymphocytes. Complement activation at the cell surface leads to lysis of the lymphocytes. Antibody reactivity against the cells (and presumably against HLA antigens) is determined by examining for the presence of dead or dying cells with the use of a vital stain such as trypan blue or eosin which can leak into cells that have holes created by complement activation. IgM antibodies have the strongest complement activating ability due to its multimeric form. IgG antibody when clustered in sufficient numbers also promote complement activation. With the use of anti-human globulin antibodies, these antibodies would bind and increase the number of antiglobulins bound to the cell surface and promote complement activation. This increase in antibodies on the surface of the cell would only occur in the presence of previously bound IgG antibodies.

Continuation of case:

One month later a repeat quick cytotoxic PRA with AHG enhancement was 96% reactive w/o DTT and 89% w/DTT. However, an ELISA using pooled HLA class I and class II antigens coated to the bottom of the wells of a 96-well plate showed only minimal reactivity. There was 4% reactivity to class I antigens and 1% reactivity to class II antigens. A 96 well ELISA was also performed in which individual HLA class I antigen types are coated on the bottom of the wells. This showed no reactivity (0%).

Question: Are these results still consistent with an immune-mediated rejection against the mismatched donor antigens?

Question: What are other possible reasons for a high percentage of reactivity using the quick PRA method and very low percentage of reactivity with the ELISA method?

Answer: These antibodies that caused a very high percentage of reactivity with the quick PRA (lymphocytoxic) method are probably not against HLA class I or II antigens. These antibodies are likely directed against non-HLA antigens found on the surface of lymphocytes. This conclusion is supported by the fact that the patient's serum showed no reactivity using both ELISA methods. In the ELISA assays, only HLA antigens are coated on the bottom of the wells used in both ELISA methods.

Continuation of case:

A call was made to the patient's floor to determine if the patient had received some form of intravenous immunoglobulin (IVIG) therapy such as anti-lymphocyte immunoglobulins (ATGAM, thymoglobulin). A review of the chart by the nurse showed no evidence of IVIG therapy or ATGAM therapy. However, further review showed that immediately before patient's MUD-BMT, he was given a conditioning regimen of Campath (Alemtuzumab), which is a recombinant DNA-derived humanized monoclonal antibody (Campath-1H) that is directed against the 21-28 kD cell surface glycoprotein, CD52. CD52 is expressed on the surface of normal and malignant B and T lymphocytes, NK cells, monocytes, macrophages, and tissues of the male reproductive system. The Campath-1H antibody is an IgG1 kappa with human variable framework and constant regions, and complementarity-determining regions from a murine (rat) monoclonal antibody (Campath-1G). (From Physicians Desk Reference).

Repeat quick PRA performed three months after MUD-BMT and Campath infusion showed only 11% reactivity with and without DTT.

A new sample of cells from the NMDP donor was obtained. T- and B-cell crossmatches performed with serum four months after transplant and Campath infusion showed no reactivity. This donor is now being scheduled for repeat stem collection for this patient.

DISCUSSION


Case IndexCME Case StudiesFeedbackHome