The Clinicopathological Features and Staging at Presentation of Gastric Cancer: A Single-Center Retrospective Study

Introduction

Gastric cancer (GC) accounts for 5.7% of total cancer incidence worldwide, which is approximately 1.0 million new cancer cases based on GLOBOCAN estimates in 2018. It is the third leading cause of cancer death (8.2%) after lung cancer and colorectal cancer, equating to 1 in every 12 deaths globally.¹ This is perhaps due to the fact that GC is often detected late, with distant metastasis at the time of diagnosis, contributing to poor prognosis.² Maconi et al³ suggested that GC is often detected at the later stage of metastasis due to nonspecific clinical presentation such as bloating and epigastric pain, while alarming symptoms such as dysphagia, anorexia, weight loss, gastrointestinal (GI) bleeding, vomiting, and anemia only occurs in minority of the patients. The study shows that the presence of, at least, one of the alarming symptoms might reduce the 5-year survival rate by an average of 26%.³ Hence, early detection in GC, prior to the emergence of such alarming symptoms, may improve the prognosis of the disease. However, time delay in GC diagnosis remains a major issue in Malaysia. A study conducted among 112 gastric and esophageal cancer patients in Malaysia reported that the mean duration from the first appearance of cancer symptoms to endoscope procedure was 32.4 weeks, which is approximately 5 months.⁴ Although there is a study which suggested that diagnostic delay of 6 months for cases of uncomplicated dyspepsia may not have an effect on patient’s survival,³ but not enough evidence had been collected to show how other clinical presentations correlates to staging of the disease. It has been well-established that clinicopathological features of the tumors are important in evaluating the disease prognosis, likewise in deciding treatment regime.⁵ This study therefore aims to identify common clinical presentations of GC, features of the tumor, and cancer staging at the time of diagnosis in Penang Adventist Hospital. Meanwhile, time delay was also evaluated by identifying the duration from which the symptoms presented was first recognized by the patient to the time of diagnosis.

Materials and Methods

A retrospective review of patients diagnosed with GC at Penang Adventist Hospital, Malaysia, for the period January 2012 to December 2018 was performed. Relevant data has been extracted from patients’ case notes, including laboratory reports, oesophagogastroduodenoscopy (OGDS), and histological and imaging studies.

Results

The study involved 69 patients (n = 69) diagnosed with GC from year 2012 to 2018 at Penang Adventist Hospital. The subjects included in the data analysis had a mean age of 62 years (SE = 1.679, median age = 62; range = 28–87) at the point of diagnosis. There were more male than female patients, with 42 males (60.9%) and 27 females (39.1%). Most patients were Chinese (60.9%), followed by Indonesian (36.2%), while Indians and other races constituted a minimal portion of this cohort, which is only 1.4% in each group (Table 1). Mean body mass index (BMI) of the subjects is 23.15 (SE = 0.588). Patients with blood group O comprised the majority of the patients, 42.6%, followed by blood group A (29.8%), B (17.0%), and blood group AB being the least (10.5%).

Smokers and ex-smokers are categorized as one group. There were more non-smokers (65.2%) than smokers (34.8%) among the study cohort. Meanwhile, patients who reported to be alcohol drinkers only constituted 14.5%, while most (84.1%) did not drink alcohol. Patients with GC who were also afflicted with diabetes mellitus, hypertension and hypercholesterolemia were 14.5, 30.4, and 14.5%, respectively, of the total sample size (Table 1).

Discussion

The results of this study show a male preponderance (60.9%) in GC patients, with mean age of 62 years. This supports global epidemiological studies, where GC was diagnosed in males 2.2 times more compare with females, which may due to protective effect of estrogen against GC.⁶ This idea was further enhanced by the presence of estrogen receptor-β in GC tumor that may protect against invasiveness in GC.⁷ Most patients were Chinese and Indonesian, who are the primary service users at the Penang Adventist Hospital. There were a larger proportion of patients who had blood group O (42.6%) compares with blood group A (29.8%), This contradicts with most studies which suggest that blood group A will have higher risk of GC than blood group O. A study suggested that although GC is more commonly associated with patients with blood group A, GC in blood group O is associated with increased risk of peptic ulcer disease, which could eventually increase the risk of GC.⁸ Therefore, this finding suggests that the development of GC among most patient in this study cohort may be associated with peptic ulcer disease or other risk factors.

Furthermore, there were only small number of patients (25.5%) who tested positive for H. pylori infection in this study, which was known to be one of the main risk factors for GC and peptic ulcer disease.⁶ These data suggested other etiologies that may have contributed to the development of GC among the Penang population. Although both smoking and alcohol drinking are known risk factors for GC,⁶ the results found in this study shows unexpected demographics, with a higher proportion of nonsmokers (65.2%) and nonalcoholics (84.1%). This may be due to self-report bias among the study cohort, where subjects would tend to deny smoking and alcohol consumption when presented in the clinic, or referred themselves as nonsmokers or nonalcoholics due to the recent halt of the consumption of these substances when they noticed their symptoms. Otherwise, this could be another evidence that suggests other etiologies which contributed to the development of GC among the study cohort, which require further studies in the future. Yusefi et al⁹ pointed out that dietary habit is one of the most important risk factors in the incidence of GC in comparison to the 52 risk factors that were reviewed. Recent local studies have shown high-salt intake,^10,11 and lack of vegetables and fruits¹² among the general population in Indonesia and Malaysia could be the main contributing factor of GC development in this study cohort. However, since this was not the main focus of the research, and most GC studies conducted in Malaysia often excluded Penang due to its unique population,² further studies involving a larger sample size, focusing on understanding GC etiology among the Penang population, is needed to support this finding.

The top three complaints presented among the study cohort were dyspepsia/ingestion (47.8%), weight loss (43.5%), and nausea/vomiting (33.3%), with mean time delay of 3.7 months. All the symptoms presented are nonspecific, making early detection of GC more challenging. According to Tata et al,⁴ the common reasons behind the delay of detection was due to the use of self-medication, empirical treatment for dyspepsia using antacid and H2 antagonist, as well as delayed endoscopy procedures in public health facilities. While endoscopic procedure delay may not be applicable in private healthcare settings in Malaysia, the tendency of the general population to self-medicate and opt for empirical treatment remain relevant,⁴ even among patients who are receiving conventional treatment.¹³

Moreover, most patients present with only one to three symptoms, suggesting the importance of implementation of GC screening; also, general practitioners should have a greater level of suspicion among high-risk patients to diagnose GC. Primary care should take note of patient’s symptoms and suspect GC in high-risk patient group, even though patient does not exhibit alarming symptoms. This is evident in countries such as Japan and Korea, where a decrease of incidence and mortality of GC was observed through the implementation of GC screening.^14,15 Howbeit, the cost-effectiveness of implementing such screening in other countries with relatively lower incidence rate remains debatable¹⁶ and the subsequent issue of overdiagnosis through screening program should be carefully addressed.¹⁷ Although there is no significant difference between the stages of cancer at presentation among patient groups that were diagnosed within the time delay of 3 months, between 3–6 months and 6 months, our study has shown that further delay of approximately one month (0.88 month) will lead to presentation of weight loss, which may be an indication of disease progression. Weight loss was known to be associated with poor prognosis, where a study has shown the significance of preoperative weight loss as an independent prognostic factor for GC, especially when weight loss is above 10%.¹⁸

In this study, most GC was found at the distal stomach (43.5%) and patients were mostly diagnosed at the later stage with poorly differentiated (67.7%) histological structure of the tumor (Tables 345). A study¹⁹ conducted in Malaysia showed that the Chinese population are more likely to have gastric adenocarcinoma located at the noncardia or distal region (73.7%), while all Chinese possessed diffuse subtype. This explains the common tumor features seen in this study, since the study cohort is largely made up of those with Chinese ethnicity and Indonesians, who may also be of Chinese origin. Regardless of the time delay and the number or types of presenting symptoms the patients reported, most patients were diagnosed at an advanced stage, where 52.6% were in Stage IV and 36.8% were in Stage III upon diagnosis. As much as 67.7% had poorly differentiated tumor, which has a poor prognosis. Meanwhile, most GC patients in this study cohort were part of the negative group for CA19.9 and CEA tumor markers, where 60% had < 35 U/mL CA19.9 reading and 72.2% has CEA tumor marker < 5 ng/mL. This could mean that CA 19.9 and CEA tumor marker studies are not suitable to be used as screening tool in GC; however other studies have shown prognostic values,²⁰ especially CEA marker in early GC,²¹ although the usefulness of both tumor markers remain debatable.²²

Conclusion

Dyspepsia, loss of weight, and nausea/vomiting were the top three symptoms presented among the study cohort. There is no evidence of specific symptom or combination of symptoms which would indicate higher risk of GC. Regardless of the number of symptoms presented by the patients and the time delay for diagnosis, most GC patients are diagnosed at the later stage of the disease, which has poor prognosis. This study therefore demonstrates the importance of development of GC screening criteria for high-risk patients based on risk factors and demographics, even prior to any symptoms presented. This is because there have been no significant presenting symptoms that could indicate an early stage of GC, neither is there a cutoff time which could be used as guideline for the general public on when to visit a clinic when a presenting symptom persisted for a period of time (i.e., time delay). Healthcare professionals may lower the threshold to perform OGDS, especially on elderly patients who have gastrointestinal-related symptoms over a span of time to ensure early detection. Further studies are required to provide guidelines for primary care practitioners on referring patients to OGDS. More importantly, there is a need to increase awareness of gastric cancer to decrease the time delay for patient to visit the clinic or being referred to a gastroenterologist.