中华骨髓库HLA分型质控工作中分型结果错误原因的分析及探讨

An analysis of HLA genotyping error results generated in the HLA quality control project of china marrow donor project

目的对中国造血干细胞捐献者资料库(简称中华骨髓库)中人类白细胞抗原(HLA)分型数据质控抽检中发现的错误分型情况进行归纳总结,探讨其产生的原因及解决策略,以提高造血干细胞捐献者HLA分型的准确性。方法采用聚合酶链反应(PCR)-序列特异性引物(sequencespecificprimers,SSP)、PCR-序列特异性寡核苷酸探针(sequencespecificoligonucleotideprobes,SSOP)及基因测序(sequencebasedtyping,SBT)分型方法,对中华骨髓库按比例随机抽取的7313份已完成HLA-A、B、DRB1基因分型的标本进行复检,使用与所抽检标本原始分型试剂不同的试剂盲检,对结果不符者,采用第3种试剂及其原始分型试剂复检;对于难以确定的结果,采用SBT方法确认。采用直接计数法进行HLA差错统计。结果质控抽检6期共7313份标本,发现HLA分型结果错误标本数183份,平均错误率为2.50%。HLA基因分型错误率逐年降低,6期错误率分别为8.18%、3.84%、2.85%、1.70%、1.10%、0.84%。HLA-A位点错误率为0.49%,其中漏检现象占A位点错误的61.11%;HLA-B位点错误率为0.85%,其中以同一宽特异性组之间的亚型判断错误者居多,占B位点错误的41.94%;HLA-DRB1位点错误率为0.66%;另外,可能由于标本搞错导致的HLA-A、B、DRB13个位点分型全错有41例,错误率为0.56%。结论错误产生的原因有实验人员的技术操作水平及工作责任心上的原因,也有分型方法及试剂本身的原因。采用DNA分型技术,使用特异性、重复性好的合格试剂及加强质控监督,有助于提高HLA分型的准确性,并将有助于提高骨髓库造血干细胞捐献者HLA分型的质量。

Objective To show the general situation of HLA typing quality control project of China Marrow Donor Project （CMDP） and probe into the causation and solving strategy, so as to improve the accuracy of HLA typing. Methods A total of 7 313 samples selected randomly by CMDP at a proper ratio were subjected to HLA-A 、B、DRB1 genotyping by polymerase chain reaction-sequence specific primers （PCR-SSP） ,polymerase chain reactlon-sequence specific oligonucleotide probes （PCR-SSOP） and sequence based typing （SBT） methods, and checked in blindness by different typing reagents. Further more, the unconformity samples were examined by the third reagent and the original typing reagents. Finally, SBT was employed to confirm the results which were hard to define. For statistics, counting method was used. Results In the checking 7 313 samples of 6 phases, 183 error samples were found, the error rate was 2.50%. The error rates of HLA gene typing were reduced year by year and the error rates of 6 phases were 8. 18% , 3.84% ,2.85% ,1.70% ,1. 10% ,0. 84% respectively. The error rate of HLA-A locus was 0.49% , and the leak examining phenomena accounted for 61. 11% A locus errors. The error rate of HLA-B locus was 0. 85% , most of them were the subtype estimation errors between same broad specificity group and accounted for 41.94% B locus errors. The error rate of HLA-DRB1 locus was 0. 66%. Besides, mistaken samples possibly result in 41 complete wrong typing of HLA-A, HLA-B, HLA-DRB1 three locus, and the error rate was 0. 56%. Conclusions The causation for these mistakes were technique operating levels of experiment personnel and their responsibility, typing methods as well as reagents themselves. To employ DNA Typing technique and qualification reagent with fine specificity and repetition and strengthen quality control surveillance will conduce to improve accuracy of HLA typing and improve quality of HLA typing.