Duregon Eleonora, Cappellesso Rocco, Maffeis Valeria, Zaggia Barbara, Ventura Laura, Berruti Alfredo, Terzolo Massimo, Fassina Ambrogio, Volante Marco, Papotti Mauro
Department of Oncology, University of Turin, 10043, Orbassano, Turin, Italy.
Department of Pathology and Cytopathology Unit, University of Padua, 35121, Padova, Italy.
Hum Pathol. 2017 Apr;62:1-7. doi: 10.1016/j.humpath.2016.09.035. Epub 2016 Dec 1.
Adrenocortical carcinoma patient prognosis is extremely variable and poorly predictable. The newly introduced Helsinki Score is the first so far proposed diagnostic and prognostic system based on the combined evaluation of morphological (mitoses and necrosis) and immunohistochemical (Ki-67) parameters. The aim of the study was to validate the prognostic role of the Helsinki Score for adrenocortical carcinoma characterization. Thus, 225 adrenocortical carcinomas were reclassified using the Weiss Score and the Helsinki Score (3× mitotic count + 5 × necrosis + Ki-67 index). At univariate analysis, statistically significant prognostic values were observed at the log-rank test for mitotic count (cutoff values: <6 and ≥55; P<.0001), Ki-67 (cutoff values: <20 and ≥50; P<.0001), Weiss Score (cutoff values: <5 and ≥8; P<.0001), Helsinki Score (cutoff values: <13 and ≥19; P<.0001), histological variant (conventional versus oncocytic; P=.009), necrosis (P=.001), and stage (P=.005). Cox multivariate analysis using a backward stepwise selection method retained only Helsinki Score and Weiss Score as predictors of poor prognosis (P<.0001 and P=.0005, respectively). Helsinki Score (with a threshold of 28.5 points; area under the curve [AUC]=0.729, 95% confidence interval=0.66-0.79) and Ki-67 (with a threshold of 20.5%; AUC=0.727, 95% confidence interval=0.66-0.79) showed the best and equivalent AUCs predicting disease-related deaths determined using receiver operating characteristic statistics. In conclusion, the Helsinki Score is a valuable system to predict prognosis in adrenocortical carcinoma, outperforming the currently established prognostic parameters.
肾上腺皮质癌患者的预后差异极大且难以预测。新推出的赫尔辛基评分是目前首个基于形态学（核分裂象和坏死）和免疫组化（Ki-67）参数综合评估提出的诊断和预后系统。本研究的目的是验证赫尔辛基评分在肾上腺皮质癌特征描述中的预后作用。因此，使用韦斯评分和赫尔辛基评分（3×核分裂计数 + 5×坏死 + Ki-67指数）对225例肾上腺皮质癌进行重新分类。单因素分析时，在对数秩检验中观察到核分裂计数（临界值：<6和≥55；P<.0001）、Ki-67（临界值：<20和≥50；P<.0001）、韦斯评分（临界值：<5和≥8；P<.0001）、赫尔辛基评分（临界值：<13和≥19；P<.0001）、组织学类型（传统型与嗜酸细胞型；P=.009）、坏死（P=.001）和分期（P=.005）具有统计学显著的预后价值。使用向后逐步选择法的Cox多因素分析仅保留赫尔辛基评分和韦斯评分作为预后不良的预测指标（分别为P<.0001和P=.0005）。赫尔辛基评分（阈值为28.5分；曲线下面积[AUC]=0.729，95%置信区间=0.66 - 0.79）和Ki-67（阈值为20.5%；AUC=0.727，95%置信区间=0.66 - 0.79）在使用受试者工作特征统计量确定疾病相关死亡的预测方面显示出最佳且相当的AUC。总之，赫尔辛基评分是预测肾上腺皮质癌预后的有价值系统，优于目前已确立的预后参数。
