TissuPath Pty Ltd. and Faculty of Medicine, Nursing and Health Sciences, Monash University School of Medicine, Melbourne, Australia.
J Urol. 2010 May;183(5):1808-14. doi: 10.1016/j.juro.2010.01.017. Epub 2010 Mar 29.
Digital image analysis software to review histopathology specimens is advancing uropathology by objectivity and reproducibility. Subjective pathologist assessed prostate tumor volume calculations correlate with known prognostic factors at radical prostatectomy. We ascertained whether image analysis software calculations of prostate tumor volume correlate with such prognostic factors, particularly positive surgical margins.
Prostate tumor volume was calculated using digital image analysis software in 851 sequential radical prostatectomy specimens. Results were correlated with clinicopathological data by logistic regression.
Median prostate tumor volume was 2.2 cc (IQR 0.9-3.8). Median peripheral and transition zone tumor volume was 1.9 and 0.6 cc, respectively. Transition zone tumors were noted in 236 specimens (27.8%) and positive surgical margins occurred in 244 (28.7%). Tumors with extensive positive surgical margins had larger image analysis software assessed prostate tumor volume (p = 0.029) and peripheral zone volume (p = 0.007) than those with only focal positive surgical margins. On univariate analysis positive surgical margin tumors were larger and had seminal vesicle invasion, extraprostatic extension, perineural invasion and higher preoperative prostate specific antigen than those without positive surgical margins (each p <0.001). A linear relationship existed between image analysis software determined prostate tumor volume, and increasing tumor stage, Gleason score and prostate specific antigen (p for trend <0.001). On multivariate analysis tumor volume and tumor volume as a percent of prostate volume predicted positive surgical margins.
Prostate tumor volume determined objectively by image analysis software correlates with positive surgical margins, as do prognostic variables such as extraprostatic extension, seminal vesicle invasion, perineural invasion, peripheral zone volume and Gleason score greater than 7. Objective digital image analysis software assessment appears to be a valid form to determine prostate tumor volume at radical prostatectomy. It is a useful adjunct to other histopathological analyses.
通过客观性和可重复性,用于评估组织病理学标本的数字图像分析软件正在推动泌尿病理学的发展。主观评估前列腺肿瘤体积的病理学家与根治性前列腺切除术时已知的预后因素相关。我们确定了图像分析软件计算的前列腺肿瘤体积是否与这些预后因素相关,特别是与阳性手术切缘相关。
使用数字图像分析软件对 851 例连续的根治性前列腺切除术标本进行前列腺肿瘤体积计算。通过逻辑回归将结果与临床病理数据相关联。
中位前列腺肿瘤体积为 2.2cc(IQR 0.9-3.8)。中位外周区和移行区肿瘤体积分别为 1.9cc 和 0.6cc。236 例标本(27.8%)中存在移行区肿瘤,244 例(28.7%)存在阳性手术切缘。具有广泛阳性手术切缘的肿瘤的图像分析软件评估的前列腺肿瘤体积(p=0.029)和外周区体积(p=0.007)大于仅具有局灶性阳性手术切缘的肿瘤。在单变量分析中,与无阳性手术切缘的肿瘤相比,具有阳性手术切缘的肿瘤更大,并且具有精囊侵犯、前列腺外延伸、神经周围侵犯和更高的术前前列腺特异性抗原(p<0.001)。图像分析软件确定的前列腺肿瘤体积与肿瘤分期、Gleason 评分和前列腺特异性抗原呈线性关系(趋势 p<0.001)。在多变量分析中,肿瘤体积和肿瘤体积占前列腺体积的百分比预测阳性手术切缘。
图像分析软件客观确定的前列腺肿瘤体积与阳性手术切缘相关,与前列腺外延伸、精囊侵犯、神经周围侵犯、外周区体积和 Gleason 评分大于 7 等预后因素相关。客观的数字图像分析软件评估似乎是一种有效的方法,可以确定根治性前列腺切除术中的前列腺肿瘤体积。它是其他组织病理学分析的有用补充。