This study was done to define clinical features for the different pathological types of advanced gastric carcinoma. One thousand one hundred three patients were identified and classified into two groups: 479 patients (43.4%) had a differentiated adenocarcinoma and 624 patients (56.6%) had an undifferentiated adenocarcinoma. Patients with the undifferentiated type were more likely to have large invasive tumors and a higher incidence of peritoneal dissemination. Conversely, the patients with the differentiated type were more likely to have a liver metastasis. Multivariate analysis, using Cox' proportional hazard model adjusted for sex, age, and other factors, suggested that tumor size was one of the seven most independent prognostic factors in patients with the undifferentiated type (relative risk = 1.01), but this parameter lost prognostic value in patients with the differentiated type. With regard to correlation between survival time and tumor size, the larger the tumor (over 10 cm), the shorter the survival time of patients with the undifferentiated type, as compared to findings in patients with the differentiated type (P less than 0.01). Thus, differences in clinical characteristics, including characteristics in the individual patients, extent of tumor, distant metastasis, prognostic factors, and prognosis correlate with the histopathological type of gastric carcinoma.

The prognosis of patients with malignancy depends upon biologic aspects of the tumour, tumour-host relationship, and the therapy prescribed. In the case of gastric cancer, some authors have reported a close relationship between the prognosis and age of the patients and others have not found a relationship (Grabiec & Owen, 1985;Mitsudomi et al., 1989). Lundegardh et al. (1986) reported that, for 34,549 patients surgically or non-surgically treated, the mortality rate for men over age 75 years was slightly higher, but there was no clear relationship, between prognosis based on clinical criteria, and age at diagnosis. Bozzetti et al. (1986) found that for patients who underwent gastrectomy survival time decreased with increasing age of the patient at the time of operation. In contrast, Coluccia et al. (1987) reported that, in patients over age 65 years, surgical resection of the gastric cancer favoured survival. In an 'early' gastric cancer, several factors are related to the prognosis (Inokuchi et al., 1983;Kodama et al., 1983;Fukutomi & Sakita, 1984;Kitaoka et al., 1984;Koga et al., 1984;Habu et al., 1986;Itoh et al., 1989). The relationship between age and prognosis due to recurrence after resection of early gastric cancer was examined herein. Adjustment for other clinicopathological prognostic factors was made using univariate and multivariate analyses. Materials and methods PatientsFor this retrospective study, we used data from 385 consecutive patients with no other simultaneous malignancy. All the patients had been treated by 'curative' gastrectomy for primary early gastric cancer, between January 1965 and December 1985 in the Department of Surgery II, Kyushu University Hospital. Early gastric cancer is defined as a lesion in which cancerous invasion is confined to the mucosa, or mucosa and submucosa, regardless of the regional lymph node metastases (Japanese Research Society for Gastric Cancer, 1981 (Dixon, 1988). The BMDP P4F and P3S programs were used in cases of the chi-square test and the Wilcoxon signed-rank test in compare groups of patients with respect to each characteristic. The BMDP PIL program was used to analyse the survival rates, using the Kaplan-Meier method, and to test for disparity of the survival curves, using the method of Mantel-Cox and the generalised Wilcoxon tests. The BMDP P2L program was used to determine which variables were independent prognostic factors for survival time by the Cox proportional hazard model in a stepwise manner (Cox, 1972). In a Cox regression analysis, age and size of tumour were included as continuous variables. Results MortalityAt the time of this analysis of data on 382 patients who underwent curative gastrectomy for early gastric cancer, the median follow-up time for 274 survivors (71.7%) was 10.2 years and 108 had died during this follow-up period (mortality, 28.3%). Of the 108 deaths, 25 were related to a recurrence of the gastric cancer (mortality, 6.5%), 18 were due to another malignancy (mortality, 4.7%) and 65 were due to another disease or to an acci...

No abstract

SUMMARYIn this paper the authors propose a structural method for a genetic algorithm (GA) for the optimization problem of cable routing in which cables have to be laid optimally. When there are no limits on the layout routes, the optimal routes for the individual cables can be found using the Dijkstra method in a finite discrete graph. However, when there are limits on the capacity in the layout routes, optimization cannot be performed using the Dijkstra method. The proposed GA is a two-level hierarchical GA that uses chromosome coding involving two levels. The routes for each cable and the combinations of cable routes are defined respectively, and layout route selection is optimized overall through genetic operations involving these levels. Block crossover and block mutation are used as genetic operations in the first level. In addition, a method to take advantage of lethal genes, which do not satisfy the constraints and are generated in the second level is used. Computer simulations confirm that the two-level hierarchical GA with these genetic operations functions effectively for the optimization problem of cable routing.

Many unsatisfied solutions being produced in applying GA to solve the constrained combinatorial optimization problems due to genetic operations. The unsatisfied solutions are regarded as lethal chromosomes in GA. Large numbers of lethal chromosomes might lead to that implementing and searching performance of GA comes to degrade. The usual means dealing with the lethal chromosomes is to eliminate it from population, however, evolved lethal chromosomes containing some fruits of evolution, abandoning lethal chromosomes is as same as abandoning available information, and leads to waste of evolving resources. We propose a new method to revive and utilize the lethal chromosomes based on immune theory, and apply it as a double islands algorithm model. To Multidimensional Knapsack Problem (MKP), simulating experiment shows that proposed method could effectively improve the performance of GA.