CARCINOMA OF THE STOMACH: A 10-YEAR SURVEY OF RESULTS AND OF FACTORS AFFECTING PROGNOSIS

Clinicopathologic characteristics of gastric cancer in a young patient population

The aim of this study was to analyze the clinicopathologic characteristics of young patients with gastric cancer with special attention to hereditary gastric cancer in a tertiary referral university hospital. Charts from all patients 40 years of age or younger at the time of diagnosis, during the period from January 1, 1987 to December 31, 2001, were retrospectively reviewed. Demographic variables, family history of gastric cancer, clinicopathologic characteristics, and treatment-related variables were analyzed. Overall survival was the main outcome variable. Survival curves were constructed by means of the Kaplan-Meier method, univariate analysis was performed with the log-rank test, and multivariate analysis with Cox regression. Significance was considered at P<0.05. During the study period, 558 cases of gastric cancer were seen at our institution, 83 (14.8%) were in patients 40 years of age or younger. Mean patient age was 33.2 years. Forty-five patients (54.2%) were male. Fourteen patients (16.9%) had a family history of gastric cancer. Five patients (6%) fulfilled the criteria of hereditary gastric cancer. Surgery was performed in 88% of patients, but only 35% of the operations had a curative intent. Operative mortality was 2.4%. On univariate analysis, advanced tumor stage, hypoalbuminemia, low performance status, diffuse type, pangastric tumor location, noncurative surgery, and lack of adjuvant chemotherapy had a significant negative impact on survival. On multivariate analysis, advanced tumor stage, pangastric tumor location, and absence of adjuvant chemotherapy were significantly associated with poor prognosis. Family history of gastric cancer or hereditary gastric cancer did not have any impact on prognosis. There is a high frequency of gastric cancer in young patients at our institution. Most patients present in advanced stages, which favors a poor overall survival. Family history of gastric cancer or hereditary gastric cancer did not have a significant impact on survival. Complete resection and adjuvant chemotherapy appeared to confer the only chance of prolonged survival.

Development of an EORTC disease-specific quality of life module for use in patients with gastric cancer

Quality of life (QL) is an important outcome in clinical trials in oncology. There is currently no valid international QL measure for gastric cancer. This paper describes the development of a QL module for gastric cancer to supplement the European Organization for Research and Treatment of Cancer (EORTC) Quality of life (QLQ-C30) questionnaire. Phases I to III of module development were conducted in the United Kingdom, France, Germany and Spain according to EORTC QL Group guidelines. Twenty relevant QL issues were generated from the literature and interviews with health professionals (n=24) and patients (n=58). This produced a 24 item provisional module. Further testing in 115 patients resulted in the QLQ-STO22, containing 22 questions, conceptualised into five scales and four single items, related to disease symptoms, treatment side-effects and emotional issues specific to gastric cancer. The use of the QLQ-C30 supplemented by the QLQ-STO22 will provide a comprehensive QL measure for international trials in gastric cancer.

Prognostic factors for gastric cancer influencing clinical practice

Despite a slow decline in the incidence of gastric cancer over the last 90 years, we can still expect to see over 100,000 patients die of this disease each decade in England and Wales. The 5-year survival rate has not improved during this century, which is largely due to the stage at diagnosis being unchanged. There are a number of prognostic determinants in gastric cancer which have clinical relevance. Age is an important determinant; patients under 40 years commonly have more advanced diffuse lesions than older patients so that a higher index of suspicion needs to be maintained in younger patients with persistent symptoms. Conflicting reports make it unclear whether the duration of symptoms bears any relationship to tumor stage, but there is some evidence that actively shortening the symptom duration by early investigation can have a beneficial effect in the proportion of patients diagnosed with early cancers. The site of the tumor is important; unfortunately, the proportion of patients with cardia lesions is increasing and this has had the effect of reducing the overall survival. Tumor size should not play a part in the decision to resect a lesion as most studies show no clear relationship between tumor size and stage. Tumor stage is the most important prognostic determinant and efforts to increase the proportion of stage I cancers presenting for surgery can be shown to alter the natural history of the disease, by diagnosing it when it is still surgically curable.

The selection of cases for randomised trials: a registry survey of concurrent trial and non-trial patients. The British Stomach Cancer Group

A randomised trial of adjuvant chemotherapy vs placebo in operable stomach cancer recruited 249 patients from the West Midlands Region between 1976-1980. A Cancer Registry survey identified a further 1261 suitable concurrent cases. Trial patients were compared with the 960 non-trial cases from participating Districts. Only 493 (51%) non-trial cases passed all of the prospective trial selection criteria for entry. Stage and fitness caused the majority of exclusions and were also highly prognostic. A univariate analysis comparing eligible patients with the trial showed the two groups to be balanced for the significant independent prognostic factors of the trial. However, differences in patient age and the surgery performed indicate that recruitment may have been influenced by unknown selection factors. This survey highlights the difficulty of retrospective selection and confirms the need for randomised controls. Data available from specialist Registries may be used to help develop new protocols and to verify and extend trial results.

Influence of "diagnostic delay" upon cancer survival: an analysis of five tumour sites

STUDY OBJECTIVE

The aim was to assess the relationship between survival, tumour stage, and the interval from first symptom to diagnosis (SDI, or duration of symptoms).

DESIGN

This was a retrospective follow up study of a cohort of patients registered in the tumour registry of the Hospital del Mar (Barcelona).

SETTING

Hospital based tumour registry, with patients derived mainly from the City of Barcelona.

PARTICIPANTS

1247 cases of lung, breast, stomach, colon, or rectal cancer were analysed using survival curves and Cox proportional hazards regression. Subjects (mean age 63.6 years) were followed for a median length of 12.9 months after diagnosis. At the time of diagnosis one fourth of patients had disseminated disease.

MEASUREMENTS AND MAIN RESULTS

Based on clinical records, a physician registered the onset time of the first symptom attributable to cancer (from which the SDI is computed), as well as the tumour stage at diagnosis. Other measurements followed standard tumour registry procedures. Overall, the crude mean SDI was 5.15 months (SD 8.03, median 2.03); only 24.5% of cases had an SDI less than a month. Crude mean SDIs by anatomical site were as follows: lung cancer 3.07 months; breast 7.44; stomach 5.34; colon 5.74; rectum 5.03. Tumour extension did not appear to be significantly influenced by SDI, only breast cancer showing a distinct pattern of increased extension with increasing SDI. As expected, the probability of survival decreased monotonically with increasing stage in all sites. Tumour site was also a significant predictor of survival, which at one year ranged from 93% for breast cancer to 28% for lung cancer. However, a longer SDI tended sometimes to be associated with a better chance of survival, a fact that was most apparent in colon cancer. All Cox proportional hazards models showed a consistent picture: SDI was not a significant predictor of survival (age and sex adjusted hazard ratios ranging from 0.97 to 1.01), neither was sex; age did predict survival, and so did site and stage.

CONCLUSIONS

The results provide further evidence of a very weak relationship between SDI and tumour stage at diagnosis (except for breast cancer), and between SDI and survival, thus emphasising some limitations within which early clinical detection operates. They also suggest that in addition to reflecting patient and physician behaviour, as well as the functioning of the health system, SDI may be influenced by the biological behaviour of the tumour.

Improving results in the treatment of gastric cancer: an 11-year audit

Of 280 patients presenting to one hospital with gastric cancer between 1975 and 1985, 97 (35 per cent) did not undergo surgery and 29 per cent (54 out of 183) of those who did had no resection performed. The 30-day operative mortality rate in the study period was 15 per cent (28 out of 183) but in the subsequent 4-year period this fell to 7 per cent (5 out of 69). The survival rate correlated significantly with depth of invasion but not with tumour site or degree of differentiation. The incidence of early gastric cancer in this series was 5 per cent but the 5-year survival rate in this group was 52 per cent suggesting that the true incidence might be even lower. The overall 5-year survival rate in our area 20 years ago was only 5.2 per cent but in this series it was 11 per cent overall and 24 per cent after resection, and with actuarial correction 15 per cent overall and 28 per cent after resection. The continuing improvement in operative mortality rates and in 5-year survival rates gives grounds for optimism, but the disease must be diagnosed earlier if this improvement in outlook is to continue.

Definition of carcinoma of the gastric cardia

This study concerns the definition of carcinoma of the gastric cardia. The topography of the esophagogastric mucosal junction (mucosal EGJ) was investigated with an endoscope in 182 patients who were free of hiatal hernias, ulcers, and neoplasms in the esophagus and stomach. The relationship between the EGJ and the cardiac gland area was then examined histologically in 56 resected specimens containing intact EGJs and cardia gland areas. Furthermore the cancerous center was determined; the shortest distance between the cancerous center and the EGJ and the amount of esophageal invasion were measured in 102 resected carcinomas located close to the junction; the carcinomas contained the EGJ and were good enough for pathohistological examination. The EGJ was located 0.5 - 1.0 cm proximal to the His angle (the gastric cardia) in radiological and endoscopic examinations. Histologically the cardiac gland area was found to straddle the EGJ at a range of about 1 cm proximal and 2 cm distal to the junction. Among the upper stomach carcinomas, most of the tumors (87.5%) whose center was located within 2 cm from the EGJ invaded the esophagus. In conclusion, carcinoma of the gastric cardia is defined as a lesion with its center located within 1 cm proximal and 2 cm distal to the EGJ.

Gastric cancer: a 25-year review

Between 1957 and 1981, 31,716 cases of gastric cancer were registered in the West Midlands, UK. The age-standardized incidence has shown a decrease from 17.42 per 100,000 population during the first quinquennium to 15.30 per 100,000 in the last. There was an apparent increase in the proportion of proximal lesions with a decrease in the proportion of distal, antral cancers. The stage of disease at diagnosis remained constant with 79 per cent of patients having stage IV disease. Less than 1 per cent presented with stage I disease. As a result, the curative resection rate was 21 per cent. The operative mortality rates for curative partial gastrectomy and total gastrectomy were 13 and 29 per cent respectively. Surgeons undertaking more than nine total gastrectomies annually had an overall mean operative mortality rate of 22 per cent. Overall age-adjusted survival at 5 years was 5 per cent. Survival at 5 years for stage I, II and III disease was 72, 32 and 10 per cent respectively. There was a significant increase in survival time for those treated by curative resection between 1972 and 1981 compared with the previous decade. The implications for the management of gastric cancer are discussed.

Lymph node involvement as the only prognostic factor in curative resected gastric carcinoma: a multivariate analysis

The aim of this study was to evaluate the independent influence of clinical and pathological variables on survival of patients with gastric carcinoma using the Cox regression proportional hazard model. Of 156 patients operated on for gastric carcinoma, 46 (29.5%) underwent palliative operation, 24 (15.5%) had a palliative resection, and 86 (55%) had a curative resection. The overall 5-year survival rate was 25 +/- 4%. After curative resection, the 5-year survival rate was 44 +/- 6%. Univariate analysis applied to these patients showed that poor survival was related (p less than 0.01) to: age (over 80 years), absence of epigastric pain, vomiting and dysphagia, total gastrectomy, tumor size (more than 4 cm), lymph node involvement (LNI), invasion through the muscularis propria, absence of intestinal metaplasia near the tumor, and linitis plastica. In multivariate analysis, lymph node involvement was found to be the only independent prognostic factor. The 5-year survival rate was 75.5 +/- 8% without LNI, 28 +/- 10% with proximal LNI, and 7 +/- 6% with distal LNI. Our results suggest that classification into 3 LNI groups is the best staging system for curative resection in gastric carcinoma.

Palliative surgery for gastric cancer

Most patients with gastric carcinoma have a disease that is too advanced for radical surgery. A Review was made of 13,175 cases of gastric carcinoma registered at the Birmingham Cancer Registry during the period of 1960-1969. Of the patients, 79.6% had disease that was not radically resected, and few of these patients survived to 2 years. Those who had a palliative resection or bypass had the lowest 30-day mortality rate when compared to all other palliative measures (P less than 0.001). Furthermore, palliative resection gave the best survival in the presence of both locally advanced and metastatic disease (P less than 0.001). This suggests that the best palliative procedure for those with a disease unsuitable for radical surgery is a resection.

Gastric cancer--delay in diagnosis and its causes

Over a period of 1 year 83 patients, admitted to Nottingham hospitals with gastric neoplasms, were interviewed in order to identify the extent of delay in diagnosis and the possible causes. The mean age was 71 (S.D. +/- 10) and there was a male preponderance of 1.8:1. The median delay from onset of symptoms to diagnosis was 22 weeks (IQR 14-35). Delay by the patient after the onset of symptoms before seeking medical help was 4 weeks (IQR 2-12). Family doctor delay was 7 weeks (IQR 3-14) and this was caused by trial of medication and radiological investigations. The hospital delay of 3 weeks (IQR 2-7) was due to patients waiting for multiple out-patient investigations, inadequate investigation of iron-deficiency anaemia, failure to follow-up gastric ulcers and difficulty in getting histological confirmation of clinically suspicious lesions. Seventeen (20%) patients were treated with a H2-receptor antagonist (Cimetidine). There was no significant difference in the delay caused by Cimetidine when compared with that due to antacids (Mann-Whitney U = 232, P greater than 0.5), and there was no difference in survival between these patients and those not treated with Cimetidine (chi 2 = 1.9, P less than 0.1). In this study only one of 80 patients had an early gastric cancer, which supports the view that gastric carcinoma is asymptomatic in its early stages and mass screening of the population would be the only way to detect carcinoma at this stage in its development. Family doctor delay can be reduced by immediate referral of patients to hospital for investigation prior to commencing medication. Hospital delay can be improved by avoiding duplication of investigations, fully investigating iron-deficiency anaemia and following up gastric ulcers with endoscopy and biopsy till fully healed.

Evolutive and prognostic aspects in gastric cancer. Analysis of 189 cases

The cases of 189 patients treated at the Surgery Service for gastric carcinoma from January 1960 to December 1978 and followed until April 1984 were reviewed in terms of evolutive behavior. Correlations between sex, age, duration of symptoms, site of tumors, lymph node metastases, stage of tumors, operability, resectability, treatment and survival were attempted. With respect to age, of the patients not operated on, survival was longer for more advanced age; the opposite occurred for the group of patients who were operated on. The impairment of lymph nodes was decisive in shortening survival; men were affected more than women and had a worse prognosis. When the tumors were located in the proximal third of the stomach, the survival rate was longer than when they were located in the distal portion and middle third of the stomach. Patients who had experienced digestive symptoms for a long time survived longer than those with a short history. It was concluded that the main prognostic factor was resectability and, regardless of any other variables, only patients from the group submitted to gastric resection survived more than 5 years.

Factors influencing prognosis in gastric cancer

In a group of 322 patients with adenocarcinoma of the stomach, 158 underwent resection. The only 5 year survivors came from the resection group. Increasing age, lymph node metastases and increasing depth of invasion of the gastric wall were all adverse prognostic features. There was a high incidence (19%) of resected patients who had suture line involvement. In spite of this there were 5 year survivors among those patients with suture line involvement and also those with lymph node involvement. The judicious implementation of an aggressive resection policy will give patients with favourable tumours the chance of a 5 year cure without involving patients with widespread neoplasm in radical surgery. Patients who had undergone previous gastric surgery for any cause had an extremely bad prognosis. Improvement in 5 year survival rates in patients undergoing resection for gastric cancer could be attributed to the increase in the number of patients with early gastric cancer.

Incidence and prognosis of N4 node involvement in gastric cancer

We have undertaken a prospective study of the frequency and prognosis associated with N4 node metastases in gastric cancer in 136 patients referred for surgical treatment between 1976 and 1983. N4 node metastases (pre-aortic or hepatic hilar nodes) were present in 20 of 31 patients who had a laparotomy without resection (64 per cent), in 2 of 8 patients who had a 'palliative' resection in the presence of distant metastases (25 per cent) and, in 19 of 85 patients who had a 'curative' resection (22 per cent). The median survival in patients having a 'curative' resection with N4 nodes was 4.5 months which was only marginally longer than in patients having a 'palliative' resection (median survival 3 months). In view of these findings and since immediate imprint cytology can be used to detect nodal metastases at operation, involvement of N4 nodes might be a contra-indication to extensive gastric resection in non-obstructing gastric cancer.

Clinicopathological staging of gastric cancer

13 175 patients with gastric carcinomas have been staged with reference to the TNM staging system. This has confirmed the poor overall prognosis (5.5 per cent and 3.8 per cent age-adjusted survival rates at 5 and 10 years respectively). However, it has demonstrated the significant survival associated with the treatment of Stage I disease (70.0 per cent at 5 years and 63.2 per cent at 10 years). This study demonstrates that 72.9 per cent of cases are unresected and that interpretation of resectability rates must be related to the population base of the study.

Changes in the location and type of gastric adenocarcinoma

To document our impression of major changes in aspects of gastric adenocarcinoma, we reviewed and compared 62 consecutive cases from 1975 through 1978 and 31 cases from 1938 through 1942. The average age at diagnosis increased from 58 to 68 years, the male to female ratio decreased to approximately 1:1, and carcinomas composed predominantly (50% or more) of signet-ring cells (SRC) increased from 9 to 39% of the total cases. In the recent series, carcinomas with SRC (compared with those without SRC) occurred nine years earlier, were more frequent in women, were located distally, and had an infiltrative growth pattern. Carcinomas originating in the proximal stomach (cardia) were not noted in the old series but formed 27% of the recent cases. These tumors showed a male predominance, contained SRC less often, and were less commonly associated with chronic gastritis. The implications of these observations are discussed.

A prospective trial of 5-FU and BCNU in the treatment of advanced gastric cancer

This study was undertaken to investigate the efficacy of a combination of 5-FU and BCNU in the treatment of advanced gastric cancer. Patients whose cancer was recurrent following surgery or in whom surgery was deemed inappropriate were entered in the study and given the two-drug combination at six-weekly intervals until either death or clear progression of disease occurred. Five of twenty-eight patients had unequivocal regression of their disease. The early optimism for this form of therapy in advanced gastric cancer is not supported by this study.

Natural history of "early" gastric cancer: results of a 10-year regional survey

In "early" gastric cancer the depth of invasion by the primary tumour is confined to the submucosa of the stomach. Out of 13228 cases of gastric cancer notified to the Birmingham Cancer Registry during 1960-9, 90 (0 . 7%) were identifed as early gastric cancer. The crude five-year survival rate of these 90 patients was 57 . 8% (age-adjusted rate 70 . 4%) compared with 3 . 7% (age-adjusted rate 4 . 7%) for all cases and 14 . 6% (age-adjusted rate 17 . 4%) for the cases treated radically. Prognostic factors established for gastric cancer pertained equally to early gastric cancer. Most of the 90 patients had had symptoms related to the gastrointestinal tract, but in contrast to patients with advanced gastric cancer only 19% had lost weight on admission. These findings suggest that early investigation of dyspeptic symptoms would increase the detection of early gastric cancer, the most important prognostic group identified in gastric cancer.

Gastric cancer detection in gastric ulcer disease

A retrospective study has been performed of all cases of gastric ulcer diagnosed or investigated within the Endoscopy Unit of the Department of Medicine, Bristol, over a three year period (1974-76). The average length of follow-up was two years. Two hundred and sixty five cases of gastric ulcer were studied of which 37 proved to be malignant (14%). Presenting complaints of anorexia, weight loss, nausea and/or vomiting, and multiple (greater than 3) symptoms, were commoner in the malignant ulcer group. Ulcer site and the presence of coexisting duodenal ulceration were largely unhelpful in deciding the status of an ulcer. Malignant ulcers tended to be large (greater than 1 cm diameter). Radiology was highly unreliable in distinguishing benign from malignant ulcers. Visual inspection at endoscopy was more reliable, but associated with a tendency to over-diagnose malignancy. False positive biopsies were uncommon (two cases). Three cases of clinically unsuspected superficial gastric carcinoma were revealed. Repeated endoscopy and biopsy of all gastric ulcers until they are completely healed is advised.

Chemotherapy in advanced gastric cancer: a controlled, prospective, randomised multi-centre study

Seventy-six patients with advanced gastric adenocarcinoma were studied in a prospecitive, randmoised, controlled trial using vincristine, methotrexate, cyclophosphamide, and 5-fluorouracil in an initiation course and mitomycin-C with 5-fluorouracil as maintenance therapy. Thirty-seven patients were inoperable and 39 had the primary tumour resected with histological evidence of residual disease. Survival in the inoperable group was short and showed no significant difference between treated and control patients. The median survival times for treated and control groups were 9.5 and 9.0 weeks respectively. In the resected patients there was no difference in ultimate overall survival between the groups but up to 20 weeks there was a suggestion that the probability of survival in treated patients was higher (P = 0.06). The patients were well-matched and it is concluded that chemotherapy has had an early effect but that a further trial with more detailed stratification, particularly of staging and histological grade, is needed. No patient received treatment for longer than two years and unacceptable toxicity occurred in only two patients. Nausea occurred more frequently in the treated group but was short-lived and clinically manageable.

Gastric carcinoma and previous peptic ulceration

The incidence of patients with gastric carcinoma having had a previously diagnosed or suspected peptic ulcer was noted in two studies. These studies were the West Midlands Gastric Chemotherapy Trial (1974--6) and a retrospective study of patients presenting to the United Birmingham Hospitals with gastric carcinoma during the years 1958--62 and 1968--72. The incidence of patients found to have had a previous operation for duodenal ulcer or to have post-mortem evidence of one was 6.5 per cent and 5.7 per cent respectively. A radiologically proved duodenal ulcer had been identified in 5.5 per cent and 2.1 per cent respectively and the incidence of previous symptoms suggestive of a duodenal ulcer was 4.0 per cent and 5.2 per cent. These figures indicate the total incidence could be 12.97--16.0 per cent, which is much higher than previously reported. The mean time interval between operation for duodenal ulcer and the development of gastric carcinoma was much longer in patients having had a partial gastrectomy than in patients having had a vagotomy and drainage procedure. The incidence of previous benign gastric ulcers was 2.5 per cent and 0.9 per cent. Fifty-three per cent of patients having had a previous operation for duodenal ulceration were found to have an unresectable carcinoma, compared with the overall unresectable rate of 35 per cent.

Clinical correlates of resectability and survival in gastric carcinoma

The course of 201 patients with carcinoma of the stomach treated from 1962 through 1966 was followed with 97% determinacy for 10 years. The actual five year survival rate was 11%; the ten year rate was 7%. The mean duration of survival was 5.8 +/- 2.7 (S.D.) months. These results were similar to those reported for the period 1922-1926. Survival was strongly correlated with the surgeon's assessment after exploration. All patients alive after five years had operations thought to be curative, usually partial gastrectomies; the survival rate of this group was 24%. Gastroenterostomy was ineffectual palliation. Better results will require nonsurgical adjuncts, since the correlates of survival are those of minimally invasive cancer.

The treatment of carcinoma of the stomach

A personal series of 200 patients with carcinoma of the stomach have been reviewed. Approximately 30% were regarded as suitable for curative resection; approximately 30% of the patients who underwent curative excision, survived five years; and approximately 30% surviving five years died of the disease before the 10th year. There were no early endoscopically diagnosed tumours in this series.

Host response and tumour biological behaviour in the two histological types of gastric carcinoma

It appears that there is validity in categorizing gastric carcinoma into two histologic types, intestinal and diffuse. The local host tissue response in 92.5% of cases of the intestinal type of gastric carcinoma was of an exudative nature. Diffuse gastric carcinoma in 70% of cases incited a dense productive fibrosis. Pools of mucin and large number of 'signet-ring' cells were mostly encountered in the intestinal type of carcinoma. Applying Dukes' parameters the tumour was found to be more than three times more invasive in cases of diffuse carcinoma. The prognostic bearing of the two histologic types, different host tissue response, behaviour of the tumour in terms of mucous production and local extension are discussed and it is suggested that diffuse gastric carcinoma carries a worse prognosis than the intestinal type. Study of a larger series of cases and longer follow-up with controlled treatment is essential to confirm this assessment.

Fibreoptic endoscopic palliative intubation of inoperable oesophagogastric neoplasms

Palliative intubation for inoperable malignant strictures at the cardia was done on 16 occasions in 13 patients using fiberoptic endoscopy. Preliminary dilatation was performed, and the Celestin tube was mounted on an introducer and passed over a guide wire inserted with a fiberoptic endoscope. Only one death resulted from the procedure and all the 12 patients who left hospital were swallowing satisfactorily on discharge. The method provided a simple and relatively safe means of relieving dysphagia and improving nutrition.

Personal experience in the surgical management of carcinoma of the stomach

A consecutive series of 226 cases of carcinoma of the stomach has been studied. The policy of treatment has been consistent throughout and laparotomy was performed in 96-5 per cent of cases. The follow-up has been studied with regard to duration of symptoms, location and degree of spread of the tumour and different forms of operative treatment. Although Billroth I gastrectomy gave a 32 per cent 5-year survival rate for the less advanced tumours the overall survival remains, as in other series, depressingly low (8-5 per cent). Education of the public to present early would not appear to help the situation as survival was directly proportional to the duration of symptoms.

Superficial carcinoma of the stomach

Nine cases of superficial gastric carcinoma have been detected with upper gastrointestinal endoscopy in Bristol in the past two years. This contrasted with only six cases found from postoperative gastrectomy specimens examined in the previous eight years. It is often difficult to distinguish a superficial carcinoma from a benign ulcer, and endoscopic diagnosis is effective only if multiple biopsy specimens are taken. Endoscopy should also be repeated and multiple specimens taken until the lesion has healed; even malignant ulcers may heal, and any healed area that is depressed with interrupted mucosal folds should be suspected of malignancy. The endoscopic and histological appearances, the age of the patients, and the clinical behaviour of the disease resembled descriptions of the disease, principally from Japan. Superficial gastric carcinoma is probably under-diagnosed in Britain.

Indolent mucoid carcinoma of stomach

In a review of 574 cases of gastric carcinoma, 50 (8.7%) proved to be mucoid using defined microscopic criteria. Three histological types were recognized: pure signet-ring cell carcinoma (three cases); tumour of mixed pattern (41 cases); and an easily recognized, well differentiated type (six cases). This last group pursued an indolent course and had a mean survival time of nine years compared with mean survival times in the other two groups of five and 18 months respectively.

Pathology and prognosis of carcinoma of the stomach

The pathology and prognosis of 205 patients treated by total or partial gastrectomy for carcinoma of the stomach during the years 1951-5 have been evaluated. One hundred and sixteen patients (61 per cent) had a partial gastrectomy and 74 (39 per cent) had a total gastrectomy. The overall postoperative mortality was 25 per cent and this was greater for total than for partial gastrectomy. The crude survival rate was 19.4 per cent at 5 years and 11.4 per cent at 10 years. Carcinoma of the cardia had a worse prognosis than tumours at other sites. There were no 5-year survivors among patients with the linitis plastica type of carcinoma of the stomach. Tumours classified as poorly differentiated or anaplastic carcinoma had a poorer prognosis than well-differentiated adenocarcinomas. The amount of lymphocytic and plasma-cell infiltration of the tumour has a very significant effect on prognosis. Five out of the 7 patients with a heavy infiltrate of such cells survived over 5 years. Lymph-node involvement is the feature having the greatest influence on prognosis. There was a 40 per cent 5-year survival rate in the absence of node involvement and an 11 per cent 5-year survival rate with involved nodes. Patients with involved nodes surviving over 5 years had, on the average, fewer positive glands than those with involved nodes who survived for less than 5 years. No benefit follows from total rather than partial gastrectomy provided the cancer and regional lymph-nodes can be removed with the simpler procedure. Four case reports are presented to illustrate the difficulties in assessing the prognosis of gastric cancer.