Sondergaard K, Schou G
Department of Pathology, Finsen Institute, Rigs-hospitalet, Copenhagen, Denmark.
Am J Dermatopathol. 1985;7 Suppl:127-9. doi: 10.1097/00000372-198501001-00023.
Cox's multivariate regression analysis for survival data was applied to facts in 1,469 clinical Stage 1 cutaneous malignant melanomas without areas of regression in order to evaluate the prognostic value of size of margins of resection and preliminary biopsy as well as sex and age of patients when accounting for variables previously shown as independent risk factors, i.e., site of lesion, thickness of lesion, level of invasion, number of mitotic figures, ulceration, lymphocytic infiltration, and predominant type of neoplastic cell. These variables were also found to be independent risk factors together with sex of patient in the present study. In contrast, size of margins of resection, preliminary biopsies, and ages of patients were found to be without prognostic significance.
考克斯对生存数据的多变量回归分析应用于1469例临床I期无消退区域的皮肤恶性黑色素瘤病例，以评估切除边缘大小、初步活检以及患者性别和年龄的预后价值，同时考虑到先前已被证明为独立危险因素的变量，即病变部位、病变厚度、浸润深度、核分裂象数量、溃疡形成、淋巴细胞浸润以及肿瘤细胞的主要类型。在本研究中，这些变量连同患者性别也被发现是独立危险因素。相比之下，切除边缘大小、初步活检以及患者年龄被发现无预后意义。
