Mao-Chih Hsieh

Surgical mortality, survival, and quality of life after resection for gastric cancer in the elderly.

Abstract. Although there were some studies on clinicopathologic characteristics, operative morbidity, and mortality in elderly patients with gastric cancer, no reports have specifically focused on survival and quality of life after resection. A total of 433 patients aged ≥ 65 years (1987–1994) who underwent gastric resection for gastric adenocarcinoma were studied. Two groups were considered: patients aged 65 to 74 years and those > 74 years. Most of the patients (78.1%) had advanced diseases, and nearly half (41.3%) had associated chronic disease(s). Resections with curative intention were performed in 362 patients (83.6%). The overall operative morbidity rate was 21.7% and mortality rate 5.1%. Although operative procedures were similar in both groups, patients aged >74 years had a higher mortality rate than those aged 65 to 74 years (10.1% vs. 3.5%; p= 0.034). Age and extent of gastric resection were two independent factors negatively affecting mortality. The cumulative survival rates for patients who underwent curative resection were 86.2%, 72.4%, 67.2%, 62.9%, and 60.0% at 1, 2, 3, 4, and 5 years, respectively. Nearly all patients (96%) after surgery had normal work and daily activities. Some patients appeared to lack energy (16%) or experienced a period of anxiety or depression. There was no statistical difference in survival and quality of life assessed by the Spitzer index after curative resection between the two groups. Therefore resection with curative intention can be performed for the elderly with acceptable morbidity and mortality rates, possible long-term survival, and good quality of life, but a limited operation should be considered in the very elderly patients.

Prognostic indicators for survival after curative resection for patients with carcinoma of the stomach

This study aims to determine prognosticindicators among patient-, tumor-, and treatmentrelatedfactors of gastric cancer patients. A total of 510patients who underwent curative gastric resection were studied. Univariate analysis of patient-relatedfactors showed a significantly lower survival inpatients with a history of obstruction, hypoalbuminemia,and anemia. Tumor-related factors including gross appearance, location, and size of tumor; depthof cancer invasion; level, number, and frequency oflymph node metastasis; stromal reaction and tumor growthpattern; and histological classification allsignificantly affected survival. Surgical treatment relatedfactors such as total or distal subtotal gastrectomy,extent of lymphadenectomy, and combined resection ofadjacent organ(s) showed a statistically significant adverse influence on survival. Multivariateanalysis identified only four tumor-related factors— number of metastatic lymph nodes, depth ofcancer invasion, stromal reaction, and gross appearanceof the tumor — as independently affectingsurvival. These findings suggest that only fourtumor-related factors were prognostic indicators inpatients with gastric cancer.

Factors affecting recurrence in node-negative advanced gastric cancer.

Background and Aim:u2002 Prognostic factors of lymph node‐negative gastric adenocarcinoma after curative resection have been discussed. Recurrent pattern of advanced lymph node‐negative gastric cancer after curative resection has rarely been described.

Relationship between age and clinical characteristics of patients with gastric cancer.

The relationship between the prognosis and age of patients with gastric cancer is controversial. To evaluate whether there is a biological characteristic specific to the age of patients, we examined the clinical characteristics of patients with gastric cancer with special reference to their age. Based on a prospective database, a retrospective study of 419 patients who underwent radical gastrectomy for cure in the past 6 years was conducted. Clinical characteristics including gender, gross appearance of the tumour (Borrmanns classification, tumour location), histopathology (depth of tumour invasion, lymph node status, Laurens classification and degree of tumour cell differentiation) and TNM tumour stage were analysed in six different age groups (< 39, 40–49, 50–59, 60–69, 70–79, > 80 years). The mean age of the 419 patients was 64.6 years (range from 26–91) and the peak age incidence of gastric cancer (46.3%) was in the 60–69 year old age group. The male: female ratio was 4.6: 1 on the whole and male gender predominated at ages > 60. The proportion of diffuse type tumours (68.4%) by Laurens criteria in the young age group (< 39 yrs) decreased with age (25% in the > 80 years group; P<0.001). Similarly, the proportion of poorly‐differentiated tumours (89.5%) in the young age group (< 39 yrs) decreased with advancing age (P<0.001). These findings suggest that both diffuse type and poorly‐differentiated tumours predominate in younger patients and, without considering the factor of delay in diagnosis, may explain the poorer prognosis demonstrated in younger patients.

Clinicopathological significance of intestinal and diffuse types of gastric carcinoma in Taiwan Chinese

The present study sought to evaluate the clinicopathological features of our gastric cancer patients in terms of a modified Laurens histological classification and compared the results with those from studies from Japan and Western countries. A total of 536 consecutive patients with gastric cancer were treated surgically between December 1987 and December 1993. A clinicopathological analysis was only performed for intestinal types (IT) and diffuse types (DT) of gastric cancer because patients falling into the ‘other’ category were too small to produce meaningful data. Of 536 cases of gastric cancer, 268 (50%) were IT, 231 (43.1%) were DT and 37 (6.9%) were of the other type. The overall IT:DT ratio was 1.2. Gastric cancer patients with IT (in contrast to DT) were characterized by older age, male dominance, smaller tumours, the tumour often being located in the lower and upper third of the stomach and less peritoneal metastasis. In both sexes, the peak incidence of gastric cancer patients was in the age group 65–74 years. Men had higher proportions of IT carcinoma than women. The IT:DT ratio for men and women was 1.4 and 0.4, respectively. Although the increased ratio of IT:DT paralleled the advance of age in both sexes, the transitional age (the age at which the IT:DT ratio exceeds 1.0) in male patients (age group 51–64 years) was earlier than for female patients (age group 65–74 years). Patients with IT gastric cancer had a significantly better survival rate than those with DT. However, the type of carcinoma (IT or DT) was not an independent predictor of survival. These data indicate that gastric cancer in Taiwan Chinese has different biological behaviour in terms of IT and DT. The clinicopathological features in IT and DT carcinoma are similar to studies from Japan and from Western countries.

A double-blind randomized controlled trial of colchicine in patients with hepatitis B virus-related postnecrotic cirrhosis

The preliminary results of a prospective double-blind controlled trial of colchicine in 100 patients with hepatitis B virus-related cirrhosis are reported. The patients, 94 males and 6 females, aged 32-80, were assigned to receive either 1 mg of colchicine or an identical placebo orally on a daily basis. The duration of the follow up ranged from 15 to 51 months (median 26 months). Seventy percent had histological proof of cirrhosis. On entry, 80 patients were in Child-Pugh class A, 19 were in class B, and one was in class C. Compared to the placebo group, there was no improvement in the colchicine group after a 24-month follow up in any of the biochemistry data, for example, serum albumin, alkaline phosphatase, alanine and aspartate aminotransferase, bilirubin, and prothrombin time. The difference in the cumulative survival rates at 51 months did not reach statistical significance (p = 0.8) in either group. There was no histological improvement in 30 patients who were willing to undergo repeated liver biopsies. No trend toward improvement of the hepatic pressure gradient was observed in these patients. The serum levels of aminopropeptide of type III procollagen increased significantly in patients in both groups after 24 months of therapy (1.07 +/- 0.06 vs. 1.36 +/- 0.06 U/ml in the colchicine group, 0.93 +/- 0.09 vs. 1.25 +/- 0.07 U/ml in the placebo group; p < 0.05). In addition, neither the clinical deterioration of cirrhosis nor death was prevented in patients receiving colchicine therapy. This report indicates that colchicine has no effect in the treatment of HBV-related postnecrotic cirrhosis.

Adenocarcinoma of midstomach : clinical and pathoanatomic relation to lymph node metastases

A detailed clinical and anatomic-pathological study of lymph node (LN) metastases from adenocarcinoma of the midstomach is lacking. To such an end, we prospectively evaluated 100 patients who underwent radical gastrectomy. A total of 3,394 LNs were removed with an average of 34 per specimen (range, 10-105). Forty-seven patients (47%) had involved LNs. The most frequent LNs involved were those along the lesser curvature (38%), greater curvature (23%), left gastric artery (19%), infrapyloric (16%), above the common hepatic artery (13%), suprapyloric (11%), and right paracardia (9%). When tumors involved the upper stomach, the frequency of LN metastases along the splenic artery and in the splenic hilum was 11.8%. Tumors along the lesser curvature or on the anterior or posterior walls had splenic hilar LN metastases in 10%, whereas of tumors involving the whole circumference, 16.7% showed with metastases. Four of 33 patients (12.1%) with early cancer had LN metastasis confined to the perigastric, left gastric artery, and above the common hepatic artery. The LN metastases were closely related to the depth of cancer invasion, location, size, gross appearance, and histology of the tumor.

Heat shock and cytokines modulate the expression of adhesion molecules on different human gastric-cancer cell lines

In order to understand the expression and modulation of adhesion molecules (AMs) on the surface of different gastric cancers, we studied 4 gastric‐cancer cell lines including SC‐MI, KATO‐III, AGS and AZ‐521. The expression of E‐cadherin, integrins (β1, β2 and β3), ICAMs (1 and 2), and CD11 (a, b and c) on the cells was detected by flow cytometry. We found that E‐cadherin was only expressed on SC‐MI and KATO‐III. CD29 (β1 integrin) could be found in cells of all 4 lines. CD54 (ICAM‐1) could not be detected in AZ‐521. In contrast, CD18 (β2 integrin), CD61 (β3 integrin), ICAM‐2, CD11a, CD11b and CD11c were all absent from these cells. Heat‐shock treatment (42.5°C, 60 min) enhanced the expression of E‐cadherin, CD29 and CD54 on SC‐M1, and of CD29 on AGS. In addition, TNF‐α (50U/ml) and IL‐1β (10U/ml) modulated the expression of these AMs, like heat‐shock treatment. The increment of these adhesion molecules caused by heat shock, TNF‐α and IL‐1β stimulation on SC‐MI was also confirmed by Western blot analysis. Functionally, these treatments increased the binding between normal human mononuclear cells and SC‐M1 cells. The heat‐shock treatment could induce a significant amount of TNF‐α and IL‐1β release from SC‐MI and KATO‐III, but seemed irrelevant to the expression of AMs. These results suggest that limited adhesion molecules were expressed on the surface of different gastric cancer cells. Heat shock, IL‐1β and TNF‐α may selectively modulate the expression of these 3 molecules on some of the cells, and this is probably related to their anti‐tumor effect.

Intestinal-type tumour in resected gastric remnant cancer.

Laurens intestinal type of gastric cancer was proposed to be dependent on long‐term environmental factors and is always preceded by chronic premalignant change. A cohort study was performed and demonstrated an increased cancer risk of gastric remnant after gastric surgery for benign disease. It is generally believed that after gastrectomy the residual stomach has an environmental change and, thus, enters a neoplastic process. Based on the carcinogenic theory of intestinal‐type tumour, it would be of interest to know whether the intestinal‐type tumour is more common in gastric remnant cancer. Forty patients with gastric remnant cancer had gastrectomy in the Veterans General Hospital‐Taipei. Another 683 patients with primary gastric carcinoma underwent resection and were used as controls. The clinical characteristics, tumour stage and intestinal‐type tumour were analysed in gastric remnant cancer and were compared with the various portions of primary gastric carcinoma. Although the overall distribution of intestinal‐type carcinoma in gastric remnant (45%) was no different to that of any other portion of stomach cancer, intestinal‐type carcinoma was more common in the early stage of gastric remnant (73%) and distal stomach (73%), but not in the proximal stomach (50%), which was supposed to have the same characteristics as the gastric remnant because of identical anatomic location. More than expected, intestinal‐type carcinoma in early gastric remnant cancer together with a long incubation interval between primary surgery and later tumour occurrence were compatible with the theory of carcinogenesis of intestinal‐type carcinoma.

Pre‐operative serum levels of tissue polypeptide antigen in patients with gastric cancer

Abstract Pre‐operative serum tissue polypeptide antigen (TPA) levels were measured in cases of gastric cancer from December 1987 to December 1992. All 351 cases received gastrectomies. The clinicopathological factors were analysed. The significant factors that correlated with the elevation of pre‐operative serum TPA levels included tumour size (> 7 cm), Borrmann‐type cancers, late stages (III and IV), lymph node metastasis, hepatic metastasis and Mings expanding type cancers. Multivariate analysis showed that the tumour size (> 7 cm) and the presence of hepatic metastasis are significant factors. To clarify the relationships between gastric cancer per se and the pre‐operative serum TPA levels, we selected cases without evidence of metastasis (n= 139). The tumour size was the only significant factor when multivariate analysis was applied. Possibilities of hepatic recurrence were found in cases with high pre‐operative serum TPA levels (> 220 U/L), even radical gastrectomies were performed. A high pre‐operative serum TPA level did not display a poor survival prognosis, if the radical gastrectomy was possible. We thus concluded that: (i) elevated pre‐operative serum levels of TPA are associated with either a large size tumour (> 7 cm) or the presence of hepatic metastasis and the tumour size is the most important factor relating to the serum TPA levels; (ii) high pre‐operative serum TPA levels (> 220 U/L) may serve as indicators of later hepatic recurrence; and (iii) elevated serum TPA levels were not indicators of survival prognosis.