Michelassi F, Takanishi D M, Pantalone D, Hart J, Chappell R, Block G E
Department of Surgery, University of Chicago, IL 60637.
Surgery. 1994 Oct;116(4):804-9; discussion 809-10.
We evaluated the influence of several clinicopathologic variables on 5-year actuarial survival rate after curative resection of gastric adenocarcinoma.
Clinical characteristics were retrieved from the records of all patients who underwent gastric resection for curative intent between 1965 and 1986 at The University of Chicago Medical Center, and follow-up was obtained from our tumor registry. Pathologic characteristics were determined from a detailed review of all available histopathologic slides.
One hundred seventy-eight patients underwent a curative resection during the study period at our institution. Overall 5-year actuarial survival rate was 29%. The relationship between clinicopathologic variables and 5-year survival rate was evaluated by Kaplan-Meier survival curve construction and chi-squared analysis. Lymphatic and/or capillary microinvasion (absent vs present, p < 0.001), tumor location (antrum and body vs gastroesophageal junction, p = 0.05), local extent of disease (limited to the gastric wall versus involving adjacent organs, p = 0.003), stage (absence versus presence of lymph node metastases, p < 0.001), Lauren type (intestinal versus diffuse, p < 0.01), and Ming type (expanding versus infiltrative, p < 0.02) significantly influenced survival. When a multivariate analysis with logistic regression of 5-year survival was performed, lymphatic and/or capillary microinvasion emerged as the only statistically significant, independent prognostic factor associated with long-term survival (p = 0.039). If microinvasion was omitted from the analysis, lymph node metastases (p < 0.05) and the extension to adjacent organs (p < 0.04) became the only statistically significant variables. Multiple correlation analyses suggested that microinvasion is an early histopathologic finding that correlates with a more aggressive natural history.
Lymphatic and/or capillary microinvasion is a more powerful predictor of 5-year survival than lymph node metastases or tumor extension to adjacent organs. Correlation among clinicopathologic variables suggests that microinvasion may represent an early finding, serving as a potential marker for a biologically more aggressive tumor.
我们评估了几种临床病理变量对胃腺癌根治性切除术后5年精算生存率的影响。
从1965年至1986年在芝加哥大学医学中心接受根治性胃切除术的所有患者的记录中获取临床特征,并从我们的肿瘤登记处获得随访信息。通过对所有可用的组织病理学切片进行详细审查来确定病理特征。
在我们机构的研究期间,178例患者接受了根治性切除术。总体5年精算生存率为29%。通过构建Kaplan-Meier生存曲线和卡方分析评估临床病理变量与5年生存率之间的关系。淋巴管和/或毛细血管微浸润(无 vs 有,p < 0.001)、肿瘤位置(胃窦和胃体 vs 胃食管交界处,p = 0.05)、疾病局部范围(局限于胃壁 vs 累及相邻器官,p = 0.003)、分期(无 vs 有淋巴结转移,p < 0.001)、Lauren分型(肠型 vs 弥漫型,p < 0.01)和Ming分型(膨胀型 vs 浸润型,p < 0.02)对生存率有显著影响。当对5年生存率进行逻辑回归多变量分析时,淋巴管和/或毛细血管微浸润成为与长期生存相关的唯一具有统计学意义的独立预后因素(p = 0.039)。如果在分析中省略微浸润,淋巴结转移(p < 0.05)和向相邻器官的扩展(p < 0.04)成为唯一具有统计学意义的变量。多重相关性分析表明,微浸润是一种早期组织病理学发现,与更具侵袭性的自然病程相关。
淋巴管和/或毛细血管微浸润比淋巴结转移或肿瘤向相邻器官的扩展更能有力地预测5年生存率。临床病理变量之间的相关性表明,微浸润可能是一个早期发现,可作为生物学上更具侵袭性肿瘤的潜在标志物。