The records of 66 patients with histologically proven adenocarcinoma of the duodenum were reviewed retrospectively to determine factors influencing survival. The parameters studied were age, sex, weight loss, jaundice, anaemia, duodenal stenosis, type of surgical procedure, tumour size and location, depth of parietal invasion, presence and location of lymph node metastases, and pancreatic invasion. These factors were assessed in a group of 46 patients who underwent curative resection of the tumour; 20 patients who received palliative procedures were excluded from statistical analysis. Survival curves were established by the Kaplan-Meier method and compared by the Mantel-Haentszel test. The actuarial 3- and 5-year survival rates of patients undergoing curative resection were 59 and 45 per cent respectively. None of the prognostic factors studied influenced survival. These results indicate that resection of adenocarcinoma of the duodenum should be performed whenever possible, even in the presence of lymph node metastasis and pancreatic spread.

At present, laparoscopic cholecystectomy is the treatment of choice for gallbladder stones. The operating technique reported by most authors includes the use of four trocars. We report a group of 710 consecutive patients treated by an original three-trocar technique. The use of the fourth trocar was necessary in only 55 cases (8%). However, among 56 cases of acute cholecystitis the use of the fourth trocar was necessary in 14 cases (25%) (p < 0.01). Twenty-six laparoscopies were converted to open procedures (3.6%). Four common bile duct injuries were observed (0.5%): two of them among the 655 operations with three trocars (0.3%) and two after application of the fourth trocar at the beginning of the procedure because of dissection difficulties. Our results are similar to those using the "classic" four-trocar technique. Moreover, this technique is less expensive and allows one less scar.

No abstract

The DiFi human colorectal cancer cell line was recently established from a familial adenomatous polyposis patient with extracolonic features characteristic of the Gardner syndrome. These cells have now been propagated for 150 passages in standard culture media and vessels without feeder layers or collagen coatings. They retain features of colonic epithelial cells such as surface microvilli, secretory vesicles, and desmosomes. Cytosol of DiFi cells contains a high level (502 U/mg protein) of the mucin CA 19-9. In addition, DiFi cells produce carcinoembryonic antigen, and induce tumors in athymic mice. Cytoskeleton analysis of DiFi cells by fluorescence microscopy showed a pronounced disorganization of actin cable structure. The isozyme genetic signature of DiFi cells is unique (0.01 probability of finding the same genetic signature in a different cell line), differs from that of HeLa cells, and has expressional features seen in other colorectal cell lines. The DiFi cell karyotype is tetraploid, contains many marker chromosomes, and shows numerous episomal particles. Two copies of chromosome 18 were absent, and only a single normal chromosome 17 was found. This parallels detection of allelic losses from DiFi cell DNA at loci on chromosomes 17p and 18 using molecular (cDNA) probes. DiFi cells clearly express transcripts for the c-myc proto-oncogene, the c-myb proto-oncogene, and the p53 tumor suppressor gene. Transforming growth factor beta inhibits DiFi cell growth in soft agar and suppresses c-myc expression in these cells. The value of this cell line in the study of genetic alterations in colorectal cancer is discussed.

To determine the potential role of sex steroid hormones in the development of colorectal tumors in humans, specific androgen (AR), estrogen (ER), and progesterone (PGR) receptors were investigated in normal mucosa (NM) and in tumor (T) paired biopsy specimens from 94 patients. Androgen receptors were detected in 98% and 96% of NM and T samples, ER in 91% and 83% of NM and T biopsy samples, whereas PGR were detected only in 14% and 10% of NM and T specimens, respectively. These incidences are independent of the sex and age of the patients. They are not related to tumor localization, histologic grade, or stages of Dukes' classification. Scatchard analysis of labeled ligand binding indicated the existence of one single class of high affinity binding sites; the calculated dissociation constant (Kd) was 1.7 +/- 0.6 10(-9) molar concentration (M) for 5 alpha-dihydrotestosterone (DHT) and 0.6 +/- 0.3 10(-9) M for estradiol (E2). These values were identical in NM and T tissue for both AR and ER. The binding capacity for DHT was 148 +/- 67 and 93 +/- 43 fmol/mg of cytosol protein in NM and T tissues, respectively (P less than 0.05). The ER content was lower and similar in NM and T biopsy specimens: 19 +/- 9 and 18 +/- 10 fmol/mg protein, respectively. The PGR content was 10 +/- 4 in NM versus 17.5 +/- 6 fmol/mg protein in T specimens. It is observed that the elevated AR in normal mucosa is not related to any known function for androgens in the digestive tract. The receptor pattern observed in tumors does not support the hypothesis previously raised in the case of chemically induced colonic tumors in rodents.