TRIAL OF A PREDICTIVE ALGORITHM TO TRANSPLANT PRA PAT

由于先前的HLA抗原致敏的患者 移植和排斥、输血或妊娠经历 在找到非反应性供体之前的等待时间更长。 与未致敏受体相比， 观察 伴随着这种延迟和结果的是增加的身体 和心理病态。 我们开发了一种计算机化的预测算法（KATS）， 在识别可接受与不可接受的私人和 从每月血清筛查结果衍生的公共HLA抗原。 本质上，该算法预测I类的选择性失配 高致敏患者的抗原。 回顾性研究 验证了算法的预测能力。 在初始 回顾性分析，其中每个参与中心利用其 自己的患者小组检查6个相同的血清，共292个细胞， 但只有4个细胞的抗原与血清相同 供体 然而，利用这些预测，215个交叉匹配呈阴性非 HLA-I类抗原。 在多中心回顾性研究中， KATS程序预测阴性最终交叉配型为95.3%， 移植结果85.4%的时间。 在另一个多中心 回顾性分析，成功移植的预测准确性 也预测了高致敏患者的发病率。 相同的算法可以 可用于预测随机分配难治性患者的相容供体 血小板输注 我们现在提出改进负交叉配血的预测， 根据算法的前瞻性预测增加器官共享 从而减少了等待时间，提高了 自愿分享肾脏的中心进行移植 根据KATS的预测和改善结果。 这些结果 将与同等致敏分层的对照组进行比较 患者的中心选择不强制分享肾脏， 基础

Patients that become sensitized to HLA alloantigens by virtue of prior transplantation and rejection, blood transfusions or pregnancy experience longer waiting times before a none-reactive donor can be found. Decreased graft survival as compared to unsensitized recipients has also been observed. Accompanying this delay and outcome are the increased physical and psychological morbidity. We have developed a computerized predictive algorithm (KATS) that is based upon the identification of Acceptable versus Unacceptable private and public HLA antigens derived from the results of monthly serum screening. In essence the algorithm predicts selective mismatches for class I antigens in highly sensitized patients. Retrospective studies have validated the predictive power of the algorithm. In an initial retrospective analysis, in which each participating center utilized their own patients panel to examine 6 identical serum, a total of 292 cells were tested yet only 4 of the cells were 4 antigen identical with the serum donor. Utilizing the predictions, however, 215 crossmatch negative non- HLA class I were identified. In a multi-center retrospective study, the KATS program predicted a negative final crossmatch 95.3% and a successful transplant outcome 85.4% of the time. In another multi-center retrospective analysis, the predictive accuracy of a successful transplant in highly sensitized patients was also predicted. The same algorithm may be used to predict compatible donors for patients refractory to random platelet transfusion. We now propose to improve the prediction for a negative crossmatch and increase organ sharing bases on prospective predictions by the algorithm thereby decreasing the waiting time, increasing the rate of transplantation at those centers volunteering to mandatorily share kidneys on the basis of the KATS predictions and improving outcome. These results will be compared to a control group of equally sensitized stratified patients whose centers choose not to mandatorily share kidneys on this basis.