Shapiro Joel, Biermann Katharina, van Klaveren David, Offerhaus G Johan A, Ten Kate Fiebo J W, Meijer Sybren L, van Berge Henegouwen Mark I, Steyerberg Ewout W, Wijnhoven Bas P L, Lanschot J Jan B van
*Department of Surgery †Department of Pathology ‡Department of Public Health, Rotterdam §Department of Pathology, University Medical Center Utrecht, Utrecht ¶Department of Pathology ||Department of Surgery, Academic Medical Center, Amsterdam, The Netherlands.
Ann Surg. 2017 Feb;265(2):356-362. doi: 10.1097/SLA.0000000000001630.
We aimed to determine pretreatment pathological tumor extent in the resection specimen after neoadjuvant chemoradiotherapy (nCRT) and to assess its prognostic value in patients with esophageal cancer.
Patients with esophageal cancer, treated with nCRT plus surgery were included (2003-2011). Pretreatment pathological T-stage (prepT-stage) and N-stage (prepN-stage) were estimated based on the extent of regressional changes and residual tumor cells in the resection specimen. Interobserver agreement was determined between 3 pathologists. The prognostic performance of prepT-stage and prepN-stage was scored using the difference in Akaike information criterion (ΔAIC). PrepN-stage and posttreatment pathological N-stage (ypN-stage) were combined to determine the effect of nodal sterilization on prognosis.
Overall concordance for prepT-stage and prepN-stage was 0.69 and 0.84, respectively. Prognostic strength of prepT-stage was similar to clinical T-stage and worse compared with ypT-stage (ΔAIC 1.3 versus 2.0 and 8.9, respectively). In contrast, prognostic strength of prepN-stage was better than cN-stage and similar to ypN-stage (ΔAIC 17.9 versus 6.2 and 17.2, respectively). PrepN+ patients who become ypN0 after nCRT have a worse survival compared with prepN0 patients, with a five year overall survival of 51% versus 68%, P = 0.019, respectively.
PrepT-stage and prepN-stage can be estimated reproducibly. Prognostic strength of prepT-stage is comparable with clinical T-stage, whereas prepN-stage is better than cN-stage. PrepN+ patients who become ypN0 after nCRT have a worse survival compared with prepN0 patients. Pretreatment pathological staging should be considered useful as a new staging parameter for esophageal cancer and could also be of interest for other tumor types.
我们旨在确定新辅助放化疗（nCRT）后切除标本中的术前病理肿瘤范围，并评估其对食管癌患者的预后价值。
纳入接受nCRT加手术治疗的食管癌患者（2003 - 2011年）。根据切除标本中消退变化的程度和残留肿瘤细胞来估计术前病理T分期（prepT分期）和N分期（prepN分期）。确定3名病理学家之间的观察者间一致性。使用赤池信息准则差异（ΔAIC）对prepT分期和prepN分期的预后性能进行评分。将prepN分期和治疗后病理N分期（ypN分期）相结合，以确定淋巴结清除对预后的影响。
prepT分期和prepN分期的总体一致性分别为0.69和0.84。prepT分期的预后强度与临床T分期相似，与ypT分期相比更差（ΔAIC分别为1.3、2.0和8.9）。相比之下，prepN分期的预后强度优于cN分期，与ypN分期相似（ΔAIC分别为17.9、6.2和17.2）。nCRT后变为ypN0的prepN +患者与prepN0患者相比，生存率更差，五年总生存率分别为51%和68%，P = 0.019。
PrepT分期和prepN分期可以重复估计。PrepT分期的预后强度与临床T分期相当，而prepN分期优于cN分期。nCRT后变为ypN0的prepN +患者与prepN0患者相比，生存率更差。术前病理分期应被视为食管癌的一个有用的新分期参数，对其他肿瘤类型也可能有意义。
