乳腺浸润性导管癌HER2免疫组化检测结果判定的重复性研究

Reproducibility in evaluating immunohistochemical results of HER2 expression in breast invasive ductal carcinoma

目的探讨病理医师对乳腺浸润性导管癌HER2免疫组化（immunohistochemistry，IHC）检测结果判定的重复性及其影响因素，寻求提高判读重复性和准确性的措施。方法参照《乳腺癌HEB2检测指南》，2位乳腺专科病理医师共同读片选取56例浸润性导管癌的HER2 IHC切片作为研究对象。4位非专科病理医师独立盲法完成对该组病例IHC切片的前后两轮结果判定。对第一轮4位观察者间的判读重复性和观察者自身的判读重复性进行Kappa分析。同时将2位专科医师共同确定的判读结果作为HER2判读参考分级，4位非专科医师第一轮的判读结果分别与之进行配对Kappa分析。结果观察者间的判读重复性为中等（K=0．5939），重复性最高的级别是HER2为0（K=0．8114），最低的是HER2为2＋（K=0．4736）。判读级别简化为阳性和阴性后，观察者间的重复性明显提高（K=0．7230）。4位观察者自身的判读重复性极好（Kw=0．8219～0．9165），判读的准确性则与经验有关。结论采用标准化的检测手段、深入学习和实践现有的判读标准以及联合判读等措施应有助于提高HER2 IHC判读的重复性和准确性。

Purpose To investigate reproducibility in scoring immunohistochemical results of HER2 expression in breast invasive ductal carcinoma and to assess how to improve the reproducibility and accuracy. Methods According to the ‘ guideline recommendations for HER2 testing in breast cancer＇ , 56 immunohistochemically stained slides of HER2 were taken by two experienced breast pathologists. All slides were then reviewed and scored twice by four general pathologists independently. Interobserver and intraobserver agreements were assessed using the kappa statistics. The results of individual pathologist were compared with the consensus opinion of the two breast pathologists. Results Interobserver reproducibility was moderate （K = 0. 593 9） with the highest value for 0 group （K = 0. 811 4） and the lowest for 2 ＋ group（K=0. 473 6）. When using two categories （positive vs. negative）, the interobserver agreement in- creased （ K = 0. 723 0）. Intraobserver reproducibility was excellent （ Kw = 0. 821 9 ～0. 916 5 ）. Scoring accuracy was associated with individual experience. Conclusions Standardized scoring, people familiar with diagnostic criteria, and evaluation by multiple observers may help to improve the reproducibility and accuracy.