HLA-Cw typing by standard serological techniques is associated with a high frequency of blanks, and reliable typing reagents for several of the Cw specificities are scarce. We evaluated the PCR-SSP technique for Cw typing in 370 kidney transplant patients and 280 healthy blood donors. Serological typing of all individuals was performed in our laboratory from 1995 to 1997 using commercially available tissue-typing trays. Comparison between serological and PCR-SSP typing revealed a discrepancy rate of 33.6% (n= 94) in blood donors and 32.4%) (n=120) in kidney recipients. Incorrect antigen assignments occurred only rarely (3.6% of the blood donors and 3.2% of the kidney recipients). The vast majority of discrepancies were due to antigens that were not detected serologically. In 26 individuals no Cw antigen was detected by serological typing, whereas PCR-SSP showed 1 allele in 13 and 2 alleles in the other 13 cases. Another 269 individuals were typed serologically with one blank (presumably homozygous). Of these, only 108 were confirmed to be homozygous, whereas an additional Cw allele was found in the remaining 161 cases using the SSP technique. Most of the "missed" specificities (86.5%) were those for which serological reagents were not available (HLA-Cw*12-*17). The most commonly "missed" specificity was HLA-Cw*1203, which occurred in 13.9% of the healthy blood donors. These results indicate that serological HLA-Cw typing is insufficient for examining the clinical importance of HLA-Cw matching in transplantation. Future studies based on molecular typing should allow the proper investigation of HLA-Cw matching in kidney and bone marrow transplantation.

No abstract

DNA typing for HLA class II improves the typing quality and this was shown previously to be relevant for kidney graft survival. In this project we addressed the question whether molecular typing for HLA class I also increases the efficacy of HLA matching in kidney transplantation. 215 HLA-A,-B,-DR zero-mismatched donor/recipient pairs as defined by serological typing were selected. Retrospective HLA-A and HLA-B typing was performed both by the PCR-SSP and the PCR-SSOP method. DNA typing for HLA-A revealed discrepant results to serology in 5.7% of the donors and 2.8% of the recipients. HLA-B typing discrepancies were found in 6.6% of the donors and 5.6% of the recipients. 10.4% of the donors and 6.5% of the recipients showed either an HLA-A or an HLA-B discrepancy Nearly one-third of the HLA-A discrepancies affected A19 splits. The most common reason for HLA-A discrepancies was the erroneous assignment of serological blanks, whereas HLA-B errors were caused mainly by the assignment of incorrect specificities. DNA typing allowed the definition of HLA-A and -B split specificities in all 118 "splitable" cases for which only broad specificities were reported based on serological typing. A total of 183 DNA class I compatible transplants had a 15% higher one-year graft survival rate than 32 transplants for which DNA typing revealed a class I incompatibility