Minoo Adib MD^•, Edna Abkarshahnazar MSc

ighly sensitized renal dialysis patients presents present an enigma to most transplant programs. Not only it is difficult to find a suitable cross-matched negative donor, but it is also apparent that a kidney transplant is generally less successful than unsensitized patients.¹ The number of patients who are on the waiting list for renal transplantation is a universal problem.²

Many patients who become sensitized have HLA-specific antibodies due to previous graft failures, blood transfusions, and pregnancies.³

Humoral sensitization and antibody definition are determined by testing patient sera using lymphocytotoxicity assays against at least 60 selected lymphocyte panels from donors with known HLA-A, B type. The percentage of lymphocytes killed by the sera is referred to as %PRA (percent panel reactive antibody).

These assays are designed primarily to detect antibodies that are specific for the products of the HLA-A and B loci. Patients with %PRA activity of more than 50% are considered to be highly sensitized.² The higher the %PRA value, the more difficult it is to find a cross-matched negative donor.

The chance of a successful transplant is improved by defining the specificity of HLA antibodies and selecting a suitable kidney without the HLA antigens, corresponding to the antibodies in the patients’ sera.¹

In this study, which was performed on 30 serum samples from sensitized patients awaiting kidney transplantation, in Aliasghar Hospital, Isfahan, the epitope specificity of HLA antibodies was defined and it was observed that the majority of antibodies have specificity for definable clusters of public epitopes.

Serum samples (1 mL) were collected from these subjects and stored at –20°C, until the beginning of serum analysis experiments.

Lymphocyte panels used for screening were obtained locally from 81 HLA-typed donors including medical students and kidney donors who were referred to laboratory on different days. Each time, 3 to 5 different HLA-typed lymphocytes were chosen and each of them was added to a separate Terasaki plate containing 30 different serum samples derived from the patients prepared previously.

Donor lymphocytes were prepared from 5 mL of peripheral blood according to the Boyum method ⁴ and using Ficoll hypaque (Axis Shield. Oslo, Norway) as a separating agent.

The 30 previously frozen serum samples were thawed and 1 µL of each sample was dispensed into a marked well, in each of the 81 Terasaki microplates, and stored at –20°C until the beginning of the experiments.

HLA-A or B typing on panel cells was performed by the microlymphocytotoxicity method developed by the National Institutes of Health (NIH).⁵ The NIH technique was also used for antibody screening of patients’ sera against 81 panel cells from HLA-typed individuals.

Reactivity of each serum sample against the 81 local cell panels was evaluated visually for testing the specificity according to Rodey and Fuller.⁶ HLA antibody identification in some of the patients’ sera was difficult because of the hetero geneity of HLA antibodies, and the presence of mixtures of different antibodies in the same sera. In addition, some HLA class I antigens have similarities in their structure, and this makes cross-reactive groups (CREGs) difficult to identify. The HLA class I antigens are divided into eight major CREGs as determined in Table 1.^{7, 8}

Defined antibody reactivities of patients’ sera were assigned to one of the two categories according to apparent epitope specificity: a) antibodies directed against private specificities were defined by reactivity against a single HLA specificity and the absence of reactivity with any other member of the same CREG (e.g., a serum "anti-A2" reacting with HLA-A2-positive cells, but not with HLA-A28, or A9 cells, was assigned to this group) and b) antibodies directed against public specificities were defined by reactivity with more than one member of a CREG (e.g., a serum "anti-A2, 28" reacting with HLA-A2 and-A28-positive cells, was assigned to this group).

Collectively, 32 recognizable HLA antibodies were defined in our patients. Results of the serologic studies are summarized in Table 2.

HLA antibody specificity and definition of the CREGs that each antibody belongs to, are summa-rized in Table 3.

Patients awaiting kidney transplantation may produce anti-HLA class I antibodies in their sera due to multiple transfusions.³

These sensitized patients tend to accumulate on renal transplantation waiting lists, because the HLA antibodies in their sera, lead to positive cross- matches with many potential organ donors.⁷

Determining HLA phenotype and also the specificity of HLA-A and B antibodies in the patients’ sera help to find a negative cross-matched, and acceptable mismatch kidney donor, against which the patients do not form alloantibodies. According to the results of this cross sectional study that was performed on 30 sensitized patients with %PRA of more than 5%, seven serum samples (23.33%) had %PRA values from 5% to 20%.

%PRA values were 20% to 87% in 16 serum samples (53.33%), and 88% or more in seven (23.33%) samples.

At 88% to 100% %PRA values, no specificities could be determined in seven (23.33%) serum samples. Almost all of the reactions in this range were positive, because the patients had developed multiple specific antibodies against a large array of HLA antigens due to multiple transfusions or hyperresponsiveness of the immune system. No auto-HLA antibodies were identified in this analysis.

These studies confirm and extend further support to previous observations^{8 – 10} that the predominant HLA class I alloantibodies produced by alloimmunized patients are directed against cross reactive epitopes. These data show that the detailed specificity analysis of highly reactive sera (%PRA > 90%) do not help to define the antibody specificity.

The observation that alloimmunized indivi-duals predominantly develop antibodies to clusters of public epitopes has important implications for developing algorithms to improve donor-recipient matching in organ transplantation.