Reproducibility of histopathological subtypes and invasion in pulmonary adenocarcinoma. An international interobserver study.
Reproducibility of histopathological subtypes and invasion in pulmonary adenocarcinoma. An international interobserver study.
复制标题
DOI：
10.1038/modpathol.2012.106
复制
发表时间：
2012-12
期刊：
影响因子：
--
通讯作者：
Kuik DJ
中科院分区：
文献类型：
--
作者：
Thunnissen E;Beasley MB;Borczuk AC;Brambilla E;Chirieac LR;Dacic S;Flieder D;Gazdar A;Geisinger K;Hasleton P;Ishikawa Y;Kerr KM;Lantejoul S;Matsuno Y;Minami Y;Moreira AL;Motoi N;Nicholson AG;Noguchi M;Nonaka D;Pelosi G;Petersen I;Rekhtman N;Roggli V;Travis WD;Tsao MS;Wistuba I;Xu H;Yatabe Y;Zakowski M;Witte B;Kuik DJ
肺腺癌的组织学亚型最近已根据主要模式进行了更新，但需要重复性数据进行验证。本研究首先评估了腺癌亚型的可重复性，然后进一步评估了一组国际肺病理学家对腺癌侵袭性和非侵袭性（全鳞屑性）模式的区分。使用显微摄影图像为基础的方法进行了两个环形研究，评估肺腺癌组织学模式的选定图像。在第一项研究中，26名病理学家回顾了典型和“困难”组织学模式的代表性图像。典型模式组合（n = 94）和困难病例（n = 21）的总得分分别为2444和546。平均kappa评分（± s.d.）五证结合组和疑难组分别为0.77 ± 0.07和0.38 ± 0.14。虽然70%的观察者将12-65%的典型图像识别为单一模式，但实性最高，微乳头最低，在92- 100%的图像中实现了主要模式的识别，除了微乳头模式（62%）。对于第二项研究的入侵，确定为一个关键的问题领域，从第一项研究，28病理学家提交和审查64图像代表典型的，以及'困难'的例子。典型病例和疑难病例的Kappa值分别为0.55 ± 0.06和0.08 ± 0.02，由于对定义侵袭的术语的不同解释，相同的病理学家一致地将其细分为侵袭性和非侵袭性类别。在肺腺癌的经典形态，其中包括大多数的情况下，有良好的可重复性，在确定一个主要的模式和公平的可重复性区分浸润性原位（完全鳞片状）模式。然而，需要更精确的定义和更好的教育来解释现有的术语，以提高对纯粹原位疾病的认识，这是一个越来越重要的领域。
Histological subtyping of pulmonary adenocarcinoma has recently been updated based on predominant pattern, but data on reproducibility are required for validation. This study first assesses reproducibility in subtyping adenocarcinomas and then assesses further the distinction between invasive and non-invasive (wholly lepidic) pattern of adenocarcinoma, among an international group of pulmonary pathologists. Two ring studies were performed using a micro-photographic image-based method, evaluating selected images of lung adenocarcinoma histologic patterns. In the first study, 26 pathologists reviewed representative images of typical and ‘difficult’ histologic patterns. A total number of scores for the typical patterns combined (n = 94) and the difficult cases (n = 21) were 2444 and 546, respectively. The mean kappa score (± s.d.) for the five typical patterns combined and for difficult cases were 0.77 ± 0.07 and 0.38 ± 0.14, respectively. Although 70% of the observers identified 12–65% of typical images as single pattern, highest for solid and least for micropapillary, recognizing the predominant pattern was achieved in 92–100%, of the images except for micropapillary pattern (62%). For the second study on invasion, identified as a key problem area from the first study, 28 pathologists submitted and reviewed 64 images representing typical as well as ‘difficult’ examples. The kappa for typical and difficult cases was 0.55 ± 0.06 and 0.08 ± 0.02, respectively, with consistent subdivision by the same pathologists into invasive and non-invasive categories, due to differing interpretation of terminology defining invasion. In pulmonary adenocarcinomas with classic morphology, which comprise the majority of cases, there is good reproducibility in identifying a predominant pattern and fair reproducibility distinguishing invasive from in-situ (wholly lepidic) patterns. However, more precise definitions and better education on interpretation of existing terminology are required to improve recognition of purely in-situ disease, this being an area of increasing importance.