Helicobacter pylori Infection is Associated with an Increased Risk of Developing Squamous Cell Carcinoma of the Oesophagus. A Cross Sectional Case Control Prospective Study

Helicobacter pylori is an important causative factor in gastric carcinogenesis. However, its role in extra-gastric gastrointestinal malignancies such as oesophageal cancer is controversial. David et al. [1] Conducted a case control study in 2010 and concluded that H. pylori infection was inversely associated with oesophageal adenocarcinoma and oesophageal gastric junction adenocarcinoma but not oesophageal squamous cell carcinoma (0SCC). Welmin et al. [2] Found in their study in 2003 that H. pylori infection increased the risk of squamous cell carcinoma of the oesophagus.


Introduction
Helicobacter pylori is an important causative factor in gastric carcinogenesis. However, its role in extra-gastric gastrointestinal malignancies such as oesophageal cancer is controversial. David et al. [1] Conducted a case control study in 2010 and concluded that H. pylori infection was inversely associated with oesophageal adenocarcinoma and oesophageal gastric junction adenocarcinoma but not oesophageal squamous cell carcinoma (0SCC). Welmin et al. [2] Found in their study in 2003 that H. pylori infection increased the risk of squamous cell carcinoma of the oesophagus.
Nie et al. conducted a meta-analysis in 2014, to evaluate the relationship of H. pylori and cytotoxin-associated gene A (CagA) positive strains with esophageal neoplasm, including oesophageal adenocarcinoma and squamous cell carcinoma of the oesophagus (OSCC) and found no significant association between H. pylori infection/CagA positive strains and OSCC. They concluded that CagA -positive strains might have a positive association with OSCC in non-Asian population and an inverse association in Asian population [3,4].
Xie et al. [5] also conducted an updated meta-analysis looking at H. pylori infection and oesophageal cancer risk. They concluded that H. pylori infection is associated with increased risk of OSCC in Eastern populations and a decreased risk of oesophageal adenocarcinoma (OAC) in the overall population.
Islami and Ramanga [6], in their meta-analysis in Nov 2012, found no overall association between H. pylori and OSCC risk. They however found that when studies were classified by geographic region, there was a statistically significant association between H.pylori and OSCC in Western studies, with OR (95%CI) of 1.65(1.17-2.32). Khoshbaten et al. [7] Concluded in their study in Iran that H. pylori infection decreases the risk of OSCC and that this is not linked to a CagA positive status.
Studies of the relationship between H. pylori infection and OSCC in South Africa where the prevalence of this cancer is high are lacking. There is a need for such studies in this country as both OSCC and H. pylori infection are common. The literature both locally and internationally shows conflicting results.
It is for these reasons that we conducted a prospective case control study to evaluate the histological prevalence of H. pylori infection in 59 consecutive patients with OSCC and 215 control group seen in Gastroenterology Division of Steve Biko Academic hospital in Pretoria South Africa over a two year period, to determine the prevalence of this infection in patients with OSCC. The prevalence of H. pylori infection in Africa, ranges from 51% to 80% depending on tests used and age group, however, since the HIV epidemic that hit our country the use of anti-viral and antibiotics have increased and this is likely to have reduced the prevalence of H. pylori infection in this population. There has not been any new studies since the appearance of this epidemic. Histological prevalence being lower [8][9][10].

Methods
Consecutive patient with advanced non curable squamous cell carcinoma of the oesophagus and an age matched control group with gastroscopy indications other than cancer referred to Gastroenterology Division of Steve Biko Academic Hospital over a two year period were recruited. All patients with this cancer present late with no possibility of cure. All patients were questioned about the use of antibiotics and or proton pump inhibitors in the last 2 months.
Patients were referred to our hospital from four provinces of South Africa, Gauteng, Mpumalanga, North West and Limpopo. Patients were referred for diagnostic and/or therapeutic upper endoscopy. An informed consent was obtained from all patients for biopsies from their gastric mucosa for H. pylori testing. In patients with bleeding peptic ulcers or suspected of use of antibiotics/ proton pump inhibitors in the last 2 months, CLO test was added to histology testing to improve accuracy. In non-bleeding ulcers (control group) CLO test was used alone for H. pylori testing.
All cancers were tested by histology with Giemsa staining. At endoscopy, two additional biopsies were taken from the antrum and two from the corpus to improve accuracy of detecting H. pylori. These biopsies were then examined by an experienced histopathologist with Giemsa stains for H. pylori.
The organism burden was reported as mild, moderate or severe. Intestinal metaplasia, dysplasia and malignancy in the gastric mucosa were reported as present or absent. Inflammation was reported as chronic, acute or inactive.
Result 59 male and female patients with squamous cell carcinoma of the oesophagus and 215 control patients without cancer were recruited. The age range was 28-83 years. A reliable drug history could not be obtained from most of these patients because of low level of literacy, however, a combination of histology and CLO test was used in 42 (control group) bleeding ulcers (Tables 1-3). H. pylori was found in a total of 30 (51%) of the 59 patients with squamous cell carcinoma and 46 (21%) 0f the 215 control group giving us a prevalence of 51% and 21% respectively p<0.001 χ 2 test. Relative risk 2.4 using csi command with chis square test in STATA. 95% CI (1.7-3.4). CLO test correlated (100%) well with histology (Table 4) (Figure 1).

Discussion
This study suggests that H. pylori infection may increase the risk of developing squamous cell carcinoma of the oesophagus by a factor of 2.4. The study, however, has a limitation in that most of the control group patients were tested with CLO test only because of resource limitation. The prevalence of H. pylori infection in South Africa is estimated to be between 50% and 80%, it would therefore be expected to be at least this high in control group. Multiple studies have suggested that H. pylori infection is protective against adenocarcinoma of the oesophagus, but increased the risk of developing squamous cell carcinoma of the oesophagus. There are no local studies in South Africa looking at this relationship but this study is in keeping with similar studies done elsewhere. Larger

Conclusion
The prevalence of H. pylori infection in patients with oesophageal squamous cell carcinoma in this study was found to be 51%, more than double that found in control group of 21%. The lower prevalence of H. pylori in the control group could be due to the HIV epidemic in our country. This result suggest that patients with squamous cell carcinoma of the oesophagus are 2.4 times more likely to be H. pylori positive than the control group. The relative risk ratio is 2:4. CLO test was found to be as sensitive and as specific as histology in those with bleeding ulcers.