Are Inflammatory Bowel Disease and Colorectal Carcinoma Associated with Helicobacter pylori ? A Prospective Study and Meta-analysis

Observational studies regarding the correlation between colorectal carcinoma, inflammatory bowel disease and Helicobacter pylori infection are inconsistent. The present study aims to investigate the association between colorectal adenocarcinoma (CRA) and inflammatory bowel disease (IBD) with H. pylori status in 100 patients who have inflammatory bowel disease and colorectal carcinoma was confirmed disease by histological approach. Besides, a meta-analysis was performed of published studies, to evaluate the link between H. pylori infection and an increased risk of CRC and IBD. Among 67 cases with CRA and 33 cases with IBD, 59.7% and 51.5% were H. pylori positive; respectively. In the meta-analysis, thirty-nine articles were included, involving 13 231 cases with CRC and 2477 with IBD. The pooled odds ratio for CRC and IBD was 1.16 (95%CI = 0.73-1.82) and 0.42 (95%CI = 0.32-0.56); respectively. Our meta-analysis indicates that H. pylori is not associated with CRC.


InTRODuCTIOn
Colorectal carcinoma (CRC) is the third most diagnosed cancer and the second most lethal cancer worldwide. 1,2In Morocco, colorectal cancer is classified as the first digestive cancer and remains a burden in the country, as 2484 new cases are diagnosed and account for ~14.8% of deaths annually. 4,5Additionally, it´s well known that colorectal cancer is sporadic.However, genetic and environmental risk factors are regarded as the most important. 6,7[10] Despite the long-standing associations between bacterial infection and carcinogenesis, researchers have recently highlighted, the implication of Helicobacter pylori (H.pylori) in the initiation of colorectal carcinogenesis and the progression of CRC. 11H. pylori is a well-known cause of gastroduodenal disease. 125][16][17][18][19] Nevertheless, the results were conflicting.In Morocco, there is no study linking H. pylori to colorectal adenocarcinoma and IBD.
To better evaluate the association of H. pylori infection with the risk of developing colorectal cancer, we aim to detect the presence of H. pylori in cases with IBD and colorectal adenocarcinomas (CRA). 20We also aim to update and review systematically current information regarding H. pylori in CRC and IBD.

PATIEnTS AnD METhODS histological study
It is a prospective study, conducted at the department of pathological anatomy in Mohammed VI University Hospital Center in Marrakech.The biopsies were obtained by colonoscopy in the gastroenterology department in the University Hospital Center of Marrakech and examined histopathologically at the anatomypathology department in the Arrazi hospital CHU Mohammed VI in Marrakech.This study included 100 cases (67 colorectal adenocarcinoma, 23 ulcerative colitis and 10 Crohn's disease cases) over 2 years (May 2018-May 2020).Medical and pathology records of the included cases were retrieved.The histopathological aspect of the study was performed by pathologists in accordance with the World Health Organisation (WHO) pathology and genetics (2010).After formalin fixation and paraffin embedding, the samples were sectioned and stained with Hematoxylin and eosine (H&E), and then analyzed by optical microscopy.The study protocol was approved by the local ethics committee of the Marrakech University Hospital Center.Patient consent was signed before the colonoscopy.In the case of illiterate or semiilliterate consenters the written consent was explained by the investigator.
Data such as: sex, age, macroscopic aspect, anatomical location, degree of infiltration, histological type and degree of tumor differentiation were collected prospectively from the medical records of the included cases.
The detection for Helicobacter pylori was performed by histological Stains: modified Giemsa, Warthin-Starry and immunohistochemical staining were used to detect Helicobacter pylori, as previously described. 13,21he statistical analysis was performed using the software SPSS v26.The ҳ2 test was used to evaluate the association between the presence of H. pylori and the variable collected, p <0.05 was considered statistically significant.

Meta-analysis Literature search
We followed PRISMA guidelines to conduct the meta-analysis.A systematic search was conducted from 1998 to 2019 using EMBASE, Web of Science, PubMed, and Cochrane Database.Two researchers (S.C and K.E) conducted literature searches independently, using the following terms: "H.pylori" or "Helicobacter pylori" and "Inflammatory bowel disease" or" IBD" or "Colorectal cancer" or "CRC" or "Crohn disease" or "ulcerative colitis" or "colitis".

Inclusion criteria
T h e i n c l u s i o n c r i te r i a we re : i ) observational studies including case-control studies, ii) detection of H. pylori by PCR, fast urease

Data extraction
The following information was extracted: I) the first author's name, ii) the year of publication, iii) the study design, iv) the country where the study was conducted, v) the method used to detect H. pylori, vi) diagnosis; and vii) sample size.

Statistical analysis
Calculations were carried out by generating odds ratio "OR" with their 95% CIs using a random-effects model.Assessment of heterogeneity was performed using the Chi-square test and the I 2 statistic.If I 2 statistic value is >50%, then the level of heterogeneity is considered as high.
Publication bias detection was performed by using Begg's rank correlation test and Egger's test.A two-sided P-value of less than 0.05 was considered as statistically significant.All analyses were performed using the software Rstudio version 1.3.1093(USA).

RESuLTS
As shown in Table 1, the median age of patients with IBD is 29.99±8.77years, and that of patients with adenocarcinomas is 64.5 ±15.933 years.The male to female ratios for IBD and CRA were 0.73:1 and 1.31:1, respectively.In the CRA group, the tumour is frequently located in the left colon and the rectum (40.3% each), while 19.4% of tumours are located in the right colon.Macroscopically, in the CRA group, ulcerativeburgeoning tumours were the most common type (86.5%).The Lieberkuhnian adenocarcinoma was the most common histologic type (98.5%), followed by Mucinous CRA (1.5%).And according to the degree of tumour differentiation, 74.6% of adenocarcinomas are moderately differentiated, 19.4% are poorly differentiated and 5.9% are well differentiated.Depending on the degree of locoregional invasion, 97% of adenocarcinomas are infiltrated into the sub-serosa followed by 1.5% that is infiltrated into the serosa.
S p e c i a l s t a i n i n g s a n d t h e immunohistochemistry revealed that H. pylori was present in 59.7% of CRA and 51.5% of IBD (Fig. 1, Table 2).41.3% of women with CRA were H. pylori positive vs 73.6% of men (p≤0.05).Also, a positive association was found between H. pylori presence and the anatomical location of the diseases and the macroscopic aspect of the tumour.No association was found between H. pylori positivity and age, histological type, degree of differentiation and infiltration of the tumour   (Table 2).
For the meta-analysis, a flow chart of study selection is reported in Fig. 2. The initial search identified 650 articles.Of these, 39 articles fulfilled the inclusion criteria and were retrieved for detailed evaluation.Twenty of these studies included 13231 patients with CRC, while nineteen articles were about 2477 patients with IBD.
In the included articles (Table 3), 22 were performed in Europe, 12 in Asia, 3 in America and two in Turkey.In terms of H. pylori detection methods, 24 studies used serological tests (ELISA), 8 used C-urea breath tests, 8 performed histological techniques (IHC, special staining), 3 used H. pylori culture used PCR.A combination of 2 to 3 detection techniques was used in 3 studies.The overall meta-analysis revealed no significant association between H. pylori and CRC (OR 1.16, 95%CI 0.73 to 1.82, p-value 0.74), and a negative association was found between H. pylori and IBD (OR 0.42, 95 % CI 0.32 to 0.56, p-value ≤0.0001) (Fig. 3, 4).However, heterogeneity was observed (p < 0.0001, I2 = 95%) (p < 0.0001, I2 = 69%) in CRC and IBD; respectively.As shown in Fig. 5, the funnel plots of publication bias appears asymmetric.Thus, we can assume the possibility of publication bias.

DISCuSSIOn
CRC accounts for 8% of cancer deaths worldwide. 4This malignancy is asymptomatic until it reaches an advanced stage. 58Nowadays, it is known that IBD has a high relationship with CRC. 59The pathogenesis of CRC and IBD is still under debate.][61] Brackmann et al., 18 mentioned that patients with CRC related to IBD are affected at a younger age.In our study, the median age of patients with IBD was 29 years old.Besides, the median age in patients with CRC was 64.5 years in the present study.A comparable result was reported in a retrospective study conducted in Rabat. 4RC is influenced by sex and gender, with males having significantly higher mortality rates. 60,61This might be due to several behavioural and gender-related factors such as a diet with red meat, alcohol consumption, and smoking.In our study, 57% of the cases were male.Depending on the CRC anatomical position, the disease progression and the overall survival of CRCs will differ.The difference between these tumours is due to different cancerogenic factors and to the developmental origin of the tumour. 62Besides, a slight decrease in the incidence of the right-sided CRC was observed worldwide That can be explained by the progress of diagnostic, treatment and by the prevention of these cancers by ablation of the adenomatous polyps in the right part of the colon.An increase was reported in the left colon CRC. 9 In a study conducted in Morocco, 60.3% of tumours were located in the rectum, 23.2% were located in the left colon, and 16.5% of tumours were located in the right colon. 63In our study, 19.4% of tumours are right-sided, 40.3% are left-sided and 40% are located in the rectum.
Regarding a probable correlation between colorectal carcinoma and H. pylori infection, several mechanisms have been proposed; such as the increasing release of gastrin that acts as a mitogen, the changing of gut microbiota and IBD induced during the migration of H. pylori from the mucosa to the light of the colon by faecal excretions. 11,64,65Besides, H. pylori virulence factors, like Cag A and Vac A that are associated with gastric adenocarcinoma, might have the same effect on CRC.In addition, an in-vitro study demonstrated that H. pylori lipopolysaccharides, can intervene with the DNA repair system of the colonic epithelial cells, promoting genotoxicity and then colon carcinogenesis. 67Also, it was shown that H. pylori lipopolysaccharides induce the production of nitric oxide, by inhibition of DNA repair enzymes and pro-apoptotic effector proteins resulting from the nitrosylation of their tyrosine and cysteine residues, causing chronic inflammation and then cancer. 68,69veral epidemiological studies have associated H. pylori infection with CRC and precancerous lesions like IBD, while others failed to establish a statistical association. 11,70herefore, a quantitative evaluation of a possible association between CRC, IBD and H. pylori is required.In the current meta-analysis, 39 studies, with 13231 CRC cases and 2477 IBD cases, fit the selection criteria.The overall analyses showed no significant association between H. pylori and CRC (OR 1.16, 95%CI 0.73 to 1.82, p-value 0.74), and a negative association between H. pylori and IBD (OR 0.42, 95% CI 0.32 to 0.56, p-value ≤0.0001).Moreover, in the present prospective study, no correlation between H. pylori and IBD was addressed.Consistently, this study has shown a negative association between H. pylori and IBD The mechanism of the protective effect is the production of IL-18 by the suppressive T cells. 71nother immunoregulatory mechanism has been proposed involving the production of H. pylori neutrophil-activating protein, that decreases inflammation by agonist ligation of toll-like receptor 2, and H. pylori DNA, that averts sodium dextran sulfate-included colitis in mice. 72owever, an association was found between sex, age, anatomical location, macroscopic aspect of the tumour in CRA patients and H. pylori (P≤0.05).These findings, plus the fact that very few studies have used PCR and histology to identify H. pylori in colorectal tissues, lead to the necessity to use more sensitive techniques to detect H. pylori in CRC subjects.
Our study presented had several limitations.First, several studies included in our meta-analysis used serological tests that can't distinguish the exact location of H. pylori.Second, few have reported the exclusion of patients who have been administrated an H. pylori eradication treatment.Third, significant heterogeneity was found across studies, which might be explained by the geographic distribution and detection methods used. 73,74

COnCLuSIOn
To the best of our knowledge, this is the first study that assesses the association between IBD and CRC with H. pylori infection in Morocco.Our results assert a possible association between H. pylori and sex, age, anatomical location and macroscopic aspect of the tumour in CRA patients.In the present meta-analysis, no association between H. pylori and CRC was established.Moreover, a negative association between H. pylori and IBD was addressed.However, more studies are needed to investigate the association of H. pylori with CRC risk using molecular techniques.

Fig. 3 .
Fig. 3. Forest plot for the pooled OR of H.pylori infection and CRC.

Fig. 5 .
Fig. 5. Funnel plots of the published studies evaluating the association between H.pylori infection and the risk of CRC (A) and IBD (B).

Table 1 .
Clinical and histopathological characteristics of patients with CRA and IBD (N= 100) CRA: colorectal adenocarcinoma, IBD: Inflammatory bowel disease.