Hans J. Altermatt

Anastomotic recurrence after resection of colorectal carcinoma has been attributed to insufficient clearance, migration of tumor cells into lymphatics, or implantation of exfoliated malignant cells during anastomosis. We studied 10 patients submitting to low anterior resection for cancer 6 to 16 cm (mean, 12.6 cm) from the anal verge. The anastomosis was performed with a circular stapler introduced transanally into the rectum using the established technique. No lavage of the rectal stump with a cytotoxic agent was conducted before the anastomosis was performed. Having completed the anastomosis, the stapler and the doughnuts were washed with saline, which was collected for cytologic examination. The doughnuts were then examined histologically; all were tumor free. In 9 of the 10 cases, malignant cells were identified in the centrifuged saline. It may be that malignant cells collected by the stapler are implanted during anastomosis and cause subsequent anastomotic recurrence.

Of 345 patients with tuberous sclerosis complex evaluated at the Mayo Clinic from 1950 to 1989, 23 were identified as having brain tumors. In 20 of the 23, histological or clinical evidence showed the tumor to be a subependymal giant cell astrocytoma. A search of the Mayo Clinic tissue registry yielded 73 giant cell-containing astrocytomas and intraventricular gliomas exclusive of ependymomas. Reexamination revealed no further examples of subependymal giant cell astrocytoma in patients without features of the tuberous sclerosis complex. Considerable histological variation was observed in the 15 subependymal giant cell astrocytomas subjected to critical microscopic review. It is of note that no correlation was noted between either the histological features, such as atypia, mitoses, endothelial proliferations, necrosis, or the flow cytometric characteristics and the clinical course or the survival time of the patients.

BACKGROUND: Hypoxia-inducible factor 1 alpha (hif-1alpha) furnishes tumor cells with the means of adapting to stress parameters like tumor hypoxia and promotes critical steps in tumor progression and aggressiveness. We investigated the role of hif-1alpha expression in patients with node-positive breast cancer. METHODS: Tumor samples from 77 patients were available for immunohistochemistry. The impact of hif-1alpha immunoreactivity on survival endpoints was determined by univariate and multivariate analyses, and correlations to clinicopathological characteristics were determined by cross-tabulations. RESULTS: hif-1alpha was expressed in 56% (n = 43/77) of the patients. Its expression correlated with progesterone receptor negativity (P = 0.002). The Kaplan-Meier curves revealed significantly shorter distant metastasis-free survival (DMFS) (P = 0.04, log-rank) and disease-free survival (DFS) (P = 0.04, log-rank) in patients with increased hif-1alpha expression. The difference in overall survival (OS) did not attain statistical significance (5-year OS, 66% without hif-1alpha expression and 55% with hif-1alpha expression; P = 0.21). The multivariate analysis failed to reveal an independent prognostic value for hif-1alpha expression in the whole patient group. The only significant parameter for all endpoints was the T stage (T3/T4 versus T1/T2: DMFS, relative risk = 3.16, P = 0.01; DFS, relative risk = 2.57, P = 0.03; OS, relative risk = 3.03, P = 0.03). Restricting the univariate and multivariate analyses to T1/T2 tumors, hif-1alpha expression was a significant parameter for DFS and DMFS. CONCLUSIONS: hif-1alpha is expressed in the majority of patients with node-positive breast cancer. It can serve as a prognostic marker for an unfavorable outcome in those with T1/T2 tumors and positive axillary lymph nodes.