Institute of Pathology, Aarhus University Hospital, Aarhus, Denmark.
Histopathology. 2011 Sep;59(3):433-40. doi: 10.1111/j.1365-2559.2011.03960.x.
Total metastatic volume (TMV) is an important prognostic factor in melanoma sentinel lymph nodes (SLNs) that avoids both the interobserver variation and unidirectional upstaging seen when using semi-quantitative size estimates. However, it is somewhat laborious for routine application. Our aim was to investigate whether digital image analysis can estimate TMV accurately in melanoma SLNs.
TMV was measured in 147 SLNs from 95 patients both manually and by automated digital image analysis. The results were compared by Bland-Altman plots (numerical data) and kappa statistics (categorical data). In addition, disease-free and melanoma-specific survivals were calculated. Mean metastatic volume per patient was 10.6 mm(3) (median 0.05 mm(3); range 0.0001-621.3 mm(3)) and 9.62 mm(3) (median 0.05 mm(3); range 0.00001-564.3 mm(3)) with manual and digital measurement, respectively. The Bland-Altman plot showed an even distribution of the differences, and the kappa statistic was 0.84. In multivariate analysis, both manual and digital metastasis volume measurements were independent progression markers when corrected for primary tumour thickness [manual: hazard ratio (HR): 1.21, 95% confidence interval (CI): 1.07-1.36, P = 0.002; digital: HR: 1.21, 95% CI: 1.06-1.37, P = 0.004].
Stereology-based, automated digital metastasis volume measurement in melanoma SLNs predicts disease recurrence and survival.
在黑色素瘤前哨淋巴结（SLN）中，总转移体积（TMV）是一个重要的预后因素，它避免了使用半定量大小估计时出现的观察者间变异和单向升级。然而，它对于常规应用来说有些繁琐。我们的目的是研究数字图像分析是否可以准确估计黑色素瘤 SLN 中的 TMV。
我们对 95 例患者的 147 个 SLN 进行了手动和自动数字图像分析测量 TMV。通过 Bland-Altman 图（数值数据）和 Kappa 统计量（分类数据）比较结果。此外，还计算了无病生存率和黑色素瘤特异性生存率。每位患者的平均转移体积为 10.6mm³（中位数 0.05mm³；范围 0.0001-621.3mm³）和 9.62mm³（中位数 0.05mm³；范围 0.00001-564.3mm³），分别采用手动和数字测量。Bland-Altman 图显示差异分布均匀，Kappa 统计量为 0.84。在多变量分析中，校正原发肿瘤厚度后，手动和数字转移体积测量均为独立的进展标志物[手动：危险比（HR）：1.21，95%置信区间（CI）：1.07-1.36，P=0.002；数字：HR：1.21，95% CI：1.06-1.37，P=0.004]。
黑色素瘤 SLN 中基于体视学的自动数字转移体积测量可预测疾病复发和生存。
