Assessment of Risk and Sero-Prevalence of Helicobacter pylori Colonization among Remote Orang Asli Tribes in Peninsula Malaysia

The epidemiology of Helicobacter pylori (H. pylori) infection is related to human poverty with marked differences between developing and developed countries. Socioeconomic factors and living standards are the main determinants of the age-dependent acquisition rate of H. pylori, and consequently its prevalence. The aim of this study was to assess the risk and sero-prevalence of H. pylori colonization among Orang Asli in Peninsula Malaysia. This cross-sectional study was conducted on Orang Asli subjects in seven isolated settlements spanning across all three major tribes (Negrito, Proto Malay and Senoi) in Malaysia. Socio-demographic characteristics of the subjects were obtained through interview. Subjects were tested for H. pylori colonization based on CagA and whole cell (WC) antigen serological assays. A total of 275 subjects participated in this study. Among these subjects, 115 (44.7%) were H. pylori sero-positive with highest sero-prevalence among Negrito (65.7%). Among subjects who were H. pylori sero-positive, CagA sero positivity was also significantly higher among Negrito. The highest proportion of respondents reported to be H. pylori sero-positive was from age group 30 years old and below (57.9%), males (56.2%), Negrito (48.6%) and live in bamboo house (92.3%). The highest proportion of respondents reported to be CagA sero-positive was from age group 30 years old and below (41.4%), males (35.6%) and Negrito (48.6%). The results of this study demonstrate that H. pylori colonization can be related to age, gender, tribes and house materials and CagA sero-positive stain closely associated with age, gender and tribes.


Introduction
Helicobacter pylori (H. pylori) is a Gram-negative bacillus capable of colonizing the human stomach [1]. More than 50% of the world population is infected with H. pylori. Prevalence of H. pylori differs based on geographical, economic and ethnic characteristics of each region [2]. Chronic colonization with H. pylori is shown to be related to gastric mucosal atrophy, intestinal metaplasia and gastric cancer [1,3,4]. The probable routes of transmission are fecal-oral and oral-oral and thus the risk factors are closely related to food and personal hygiene [5]. Risk factors include low educational level, large number of siblings, household conditions, absence of plumbing system and sanitary facilities, and poor hygiene conditions [6]. H. pylori colonization may be diagnosed through invasive (rapid urease test, histology and culture of biopsy specimens) and non-invasive techniques (serological tests, fecal antigen assessment and urea breath test) [7]. The serological assessment is considered an acceptable detection technique which may be able to detect past (less than 1 year) and current colonization [7]. Serological assessment determines the presence of Immunoglobulin G (IgG) against H. pylori [7]. Moreover, serological test may provide additional data on the virulence of the H. pylori by detecting antibodies against CagA antigen [8].
Malaysia is a country with different ethnicities; among them Malay, Chinese and Indians are the largest sub-populations while Orang Asli (aborigines) comprises only 1% of the population of Malaysia. Three major tribes of Orang Asli are present in Peninsular Malaysia. These tribes include; Negrito, Senoi, and Proto-Malay [9]. The Malaysian government has relocated most of the Orang Asli to periphery of towns with basic education and healthcare facilities. However, most of the Orang Asli remain illiterate and therefore are less aware of personal hygiene and sanitation. The Orang Asli Negrito tribe, also known as Semang, is believed to be the earliest to arrive in Peninsular Malaysia about 25,000 years ago [10,11]. They were separated into six sub-tribes, which are the Kintak, Kensiu, Batek, Mendrik, Jahai, and Lanoh [12]. They are the present day descendants of the early Hoabinhians, who were largely nomadic foragers. However, they now live in permanent settlements in the central, northern and eastern region of Peninsular Malaysia [13]. The Negrito is one of several populations of the seafarer Negrito (which includes Andaman Islanders, the Aeta in Philippines and some Papuan) who are remnants of a previously widely spread Asian population [14]. The Senoi, which include the Mah Meri, Semok Beri, Temiar, Che Wong, Jah Hut, and Semai, reached Peninsular Malaysia during the second wave of migration about 8000 years ago from South Asia, the mountain areas of Cambodia, Vietnam and Burma [12,15]. The Senoi speak Austro-Asiatic languages of the Mon-Khmer sub-group, which reflects their ancient connection with the mainland Southeast Asia [12,13]. However, some believe the Senoi are descendants of Australoid from Australia and Veddoid from South India [16]. The Proto Malays consist of the Jakun, Temuan, Semelai, Kuala, Kanak, and Seletar [17]. They live mainly in the southern part of Peninsular Malaysia and are known to be similar to the Deutero-Malays not only from a morphological standpoint but also culturally and linguistically [11,18]. However, there was evidence of intermarriage and assimilation between this group and the other two earlier groups [14].
The prevalence of H. pylori colonization was reported to range between 24.3% to 49% in Malaysia as a whole based on histological assessment of samples obtained from endoscopy [4,19,20,21]. Prevalence of H. pylori was reported to be higher among Indian (61.8%) and Chinese (48.1%) ethnicities compared to Malay ethnicity (16.4%) [4,19]. Risk factors for H. pylori colonization among the general Malaysian population includes age (older than 45 years old), Chinese and Indian ethnicities, and low educational level [4,19]. These data are mostly related to the most prevalent ethnicities in Malaysia [4,19,20,21]. H. pylori colonization and the related risk factors in the Orang Asli was previously assessed in an earlier study [20]. The risk factors reported were age, sex, occupation, educational level, number of family members in each household, smoking status, alcohol consumption, herbal plants and roots use, source of water supply, diet of exotic foods and family history of gastric disease. In the study, 480 Orang Asli from a settlement in northeastern Malaysia (210 km off Kota Bahru, Kelantan), the prevalence of H. pylori colonization was reported to be 19% based on serological test for IgG [20]. These findings indicate that H. pylori colonization does not follow the common national pattern and therefore there is a need for further assessment of the prevalence of H. pylori colonization and socio-demographic characteristics of the Orang Asli from other settlements. The Orang Asli population in Malaysia is made up of different tribes and sub-tribes, which resides in different geographical regions in the country and with very different lifestyle. In addition, no data exists on the prevalence of virulent H. pylori strains in the Malaysian Orang Asli population. The main aim of this study was to assess the prevalence of H. pylori and the prevalence of the more virulent strains (based on CagA sero-positivity) in the Orang Asli population. The second aim was to investigate the association between host socio-demographic characteristics and H. pylori and CagA IgG sero-positivity.

Study Subjects
Seven sub-tribes representing the three indigenous populations from Peninsular Malaysia were included in this study namely, Senoi, Negrito and Proto-Malay (Table 1; Fig 1). A total of 257 samples were recruited for this study. Prior to sample collection, customary visits were made to meet the headmen of the sub-tribes and the community members to explain the nature of the study and to seek their consent. All the households in the sampled indigenous population were invited to participate in the interview survey. Only one member per household was randomly selected to be interviewed. Participant's informed written consent was obtained before commencement of the interview. The interview and the process of obtaining informed consent were conducted in Malay language and witnessed by an accompanied JAKOA officer. Inclusion criteria was participants aged 18 and above and originating from and living in the selected villages.
The questionnaire consist of four mains sections. The first section of the survey consists of socio-demographic characteristics which include educational level, occupation, average household income and type of house. The second section of the questionnaire consists of the activities and lifestyle of participants. This section is divided into three parts which are: a) Contact with domestic or wild animals or insects, b) Personal hygiene and c) Food and nutrition. The third section of the survey consists of questions about gastrointestinal symptoms. The last section consists of family history of H. pylori colonization.

H. pylori Serum Antibody Analysis
A total of eight millilitres of peripheral blood was collected from the participants. Serum IgG was detected by ELISAs, using either H. pylori whole cell (WC) or CagA protein as antigens. WC antigens were derived from a characterized pool of samples from five US strains. Sensitivity and specificity of this serologic test was 96% and 93.5%, respectively [22,23]. For the WC antigen, an optical density ratio 1.1 in 1:800 H. pylori was defined as sero-positive. When the ODR-value was in the range of 0.9 and 1.1 the result was considered equivocal and when the ODR value was <0.9 the result was considered sero-negative.
To assess serologic response to CagA, a 66-kDa cagA fragment that had been cloned in Escherichia coli as pORV220, was used as antigen [24]. For this assay, serum was diluted 1:100 and was considered sero-positive if the optical density was 0.35 and a sero-negative result was < 0.30. An equivocal result for CagA was defined when the ODR-value was in the range of 0.3 and 0.35.

Statistical Analysis
Data was analyzed using IBM SPSS Statistic 21.0 for Windows. The Chi-square was used to test group differences. Bivariate analysis (chi-square test) was conducted to test between sociodemographic with CagA and H. pylori sero-positivity. Significant predictors of CagA and H. pylori sero-positivity were assessed using multiple logistic regressions. The significance level was determined at p-value 0.05.

Ethical Statement
This study was reviewed and approved by the

Lifestyle Characteristics
Majority of the respondents were never or rarely swimming in river or stream (55.4%) and hunting (79.9%). However, majority of them reported sometimes and often fishing in river or stream (51.8%). Only 26.6% were reported currently smoking and majority never or rarely consumed alcohol (98.6%). In terms of having contact with domestic and wild animals, most of the respondents never have contact with cat (61.2%), dog (85.6%), bird (92.8%), goat (99.3%), pig (99.3%), wild pig (84.9%), monkey (86.3%) and wild bird (86.3%). All of the respondents reported never had contact with cow. However, most of them were reported to ever have contact with chicken and duck (51.1%). Majority of the respondents reported no pests such as flies (53.2%), cockroaches (60.4%) and mice or rodents (71.9%) in the house. Most of the respondent reported that they sometimes or often boiled water before drink (98.4%), wash hands with soap (99.2%), wash hands before meals (98.4%) and after toilet use (98.3%) as well as walk barefoot outside the house (55.3%). In contrast, majority of the respondent reported never or rarely use the same glass or cup (65.3%) and using same dish or bowl without washing it first (73.4%). The majorities reported sometimes or often eat fiber food (87.1%), poultry products (65.3%), fish or seafood (76.6%) as well as drink milk (66.1%), coffee (63.75) and tea (79.0%).
On the other hand, the majorities reported never or rarely eat red meat (62.9%), raw meat (95.2%) and raw vegetables (61.0%) ( Table 2).  Reporting pest problems in the house:

Characteristics of Gastrointestinal Symptoms
When the respondents were asked about having experienced gastrointestinal symptoms, most of the respondents reported that they never or rarely experienced abdominal pain or unpleasant sensation (77.7%), prolonged persistence of food in stomach (79.1%), feeling of fullness after starting to eat (71.2%), burning sensation (81.3%), reflux or regurgitation (82.6%), upper abdominal bloating (79.0%) and nausea (88.5%) ( Table 3).  Table 4. Among these subjects, 115 (44.7%) were found to be either positive for anti-WC and/or anti-CagA. In contrast, only 86 (33.5%) subjects were sero-positive for anti-WC and 72 (28%) were sero-positive for anti-CagA. Only 43 (16.7%) subjects were positive for both anti-WC and anti-CagA while 72 (28.0%) were positive for either anti-WC or anti-CagA. H. pylori sero-positivity was significantly higher among Negrito compared to Senoi (p-value <0.001). Among subjects who were H. pylori sero-positive, CagA sero-positivity was significantly higher among Negrito compared to Proto Malay and Senoi (p-value = 0.016 and 0.02).

Association between H. pylori Sero-Positivity with Socio Demographic, Lifestyle Characteristics and Gastrointestinal Symptoms
Univariate analysis showed that H. pylori sero-positivity exhibited significant differences across age, gender, tribes and house material. The respondent with equivocal status of H. pylori was taken out due to small group size. There were no statistical significant differences between H. pylori sero-positivity and all gastrointestinal symptoms variables. The highest proportion of respondents reported to be H. pylori sero-positive was from age group 30 years old and below (57.9%), males (56.2%), Negrito (67.0%) and live in bamboo house (92.3%). Multivariate logistic regression analysis revealed respondent who lived in wooden house (OR = 0.062, 95% CI, 0.006-0.603, p<0.05) has significantly smaller likelihood to being diagnosed as H. pylori sero-positive compared to those with a reference level of bamboo type of house. It is noted that all 16 respondents who lived in bamboo houses were of Senoi origin. Among respondents living in bamboo houses, 12 (75.0%) were H. pylori sero-positive, .05) has significantly higher likelihood of being diagnosed as H. pylori sero-positive compared to those with a reference level of brick or cement house. Respondent who lived in the brick or cement house also has smaller likelihood (OR = 0.070, 95% CI, 0.009-1.012) to being diagnosed as H. pylori sero-positive compared with those with a reference level of bamboo type of house. However, this finding is not statistically significant.
Respondents with age group of more than 30 years old (OR = 0.588, 95% CI, 0.205-1.686) were less likely to being diagnosed as H. pylori sero-positive compared to with those of reference level of age group of 30 years old and below. Women respondents were more likely (OR = 1.811, 95% CI, 0.503-6.523) to being diagnosed as H. pylori sero-positive compared with those of reference level of male. Negrito tribe (OR = 1.036, 95% CI, 0.306-3.503) and Proto Malay tribe were more likely (OR = 2.010, 95% CI, 0.329-12.257) to being diagnosed as H. pylori sero-positive compared with those of reference level of Senoi tribe. However, these findings were not statistically significant (Table 5).

Association between CagA Sero-Positivity with Socio Demographic, Lifestyle Characteristics and Gastrointestinal Symptoms
Univariate analysis showed that CagA sero-positivity exhibited significant differences across age, gender and tribes. There were no statistical significant differences between CagA sero-positivity and lifestyle characteristics variables and gastrointestinal symptoms variables. The highest proportion of respondents reported to be CagA sero-positive was from age group 30 years old and below (41.4%), males (35.6%) and Negrito (48.6%).
Multivariate logistic regression analysis revealed age group of more than 30 years old (OR = 0.472, 95% CI, 0.255-0.873, p<0.05) significantly were less likely to be CagA sero-positive compared to with those of reference level of age group of 30 years old and below. Negrito Table 5. tatistical analysis of relationship of H. pylori sero-positivity with socio-demographic data.

Details
Frequency, N (%) H. pylori status (n = 244) Multiple logistic regression for sero-positive vs sero-negative In the other hand, Proto Malay tribe was less likely (OR = 0.802, 95% CI, 0.094-6.815) to be CagA sero-positive as compared with those of reference level of Senoi tribe. Female respondent were less likely (OR = 0.687, 95% CI, 0.372-1.270) to be CagA sero-positive compared with those of reference level of male. However, there are no statistical significant differences for these findings (Table 6).

Discussion
This study provided us the opportunity to investigate the risk factors and prevalence rate associated with H. pylori colonization among Orang Asli tribes residing in less accessible areas of Peninsular Malaysia. Many of these villages were only accessible by four-wheel drive vehicles and without stable electrical supplies. Thus, serology-based assays were chosen to investigate the extent of exposure to H. pylori using sera collected from Orang Asli living in remote villages.
Whole cell antigen used in our serology test was prepared from a mix of five US H. pylori strains and previous studies have shown that antigenic differences exist among the H. pylori strains from different geographical regions [24,25]. To overcome the problem of poor antigenic recognition, two different antigens (WC and CagA) were used in this study. In a previous study in Ladakh, it was shown that as much as 27% of H. pylori culture-positive subjects responded to only the CagA antigen but not the WC antigen [24,25]. Thus, in this study, positive for WC and/ or CagA serology test was taken as H. pylori sero-positive.
Univariate analysis found that H. pylori sero-positive prevalence was higher among respondents who lived in bamboo house. In the multivariate analysis, house material is also found to be a significant predictor for H. pylori, where Senoi respondents who lived in bamboo house were at higher risk of being infected with H. pylori. To date, most Orang Asli has been relocated to the better settlement with better amenities and facilities by the government [26]. The groups who still lived in the bamboo house can be considered as primitive groups where they lived in deep of the jungle or very remote areas. Unpublished finding from a study done on Orang Asli in Peninsular Malaysia shows that majority of Orang Asli who lived in the bamboo house used well and river water for drinking, washing and bathing, which led them to be at higher risk of being infected with H. pylori. A significantly higher proportion of males than females were H. pylori sero-positive. Majority of the males respondents in this study were smokers. Smoking is reported to cause an increase in acid and pepsin secretion and changes in gastric motility, prostaglandin synthesis, gastric mucosal blood flow and mucus secretion [27].
Various studies have reported that the prevalence of H. pylori colonization increases with age [28,29,30]. In this study, the likelihood of being diagnosed as H. pylori sero-positive was higher for respondents aged 30 years old or below, however the association is not statistical significance. Nevertheless, our study found significant association age and CagA status. Further, participants of older age group who were H. pylori sero-positive were significant less likely to be CagA sero-positive compared to the younger age group. This finding was attributable perhaps to a change in H. pylori population involved in more recent incidents of colonization. The cagAstrains that had colonized older Orang Asli above 30 years old in the past may less virulent than the cagA+ strains that are colonizing the younger Orang Asli now. Unfortunately, due to limited medical facilities and equipment's on-site, it was not possible to perform endoscopic examination and to collect gastric biopsy samples for H. pylori culturing and genotyping.
In this study, H. pylori and CagA sero-positivity were found to be significantly higher in Negrito compared to Proto Malay and Senoi. The difference between the lifestyle and practices may contribute to the difference in H. pylori sero-prevalence. Likewise, tribes was also significantly associated to CagA status. Negrito have significantly greater likelihood of being colonized with cagA+ strain compared to Senoi and Proto Malay. cagA+ strains were found in 74.0% Negrito who were H. pylori sero-positive. According to Tan et al (2005), cagA+ strains are the predominant genotypes in all ethnic groups in Malaysia; Malay, Chinese and Indian [31]. Similarly, the sero-prevalence of H. pylori among asympomtatic volunteers from highly urbanized Kuala Lumpur was 38.6% (91/236) and among those positive for H. pylori by serology, 83.5% (76/91) were sero-positive for CagA (data not shown). This lead us to hypothesize that the local aborigines had low prevalence of H. pylori and were colonized mainly with the less virulent cagAstrains when they first settled in this part of the world. This view is further supported by the sequencing of the Sahul64, a H. pylori strain isolated from an indigenous Australian, which lacks the cagPAI and posseses a nontransportable VacA protein [32]. Negrito have been known to interact with Malays who resided along the fringes of the forest [26]. Hence, the recent increase in colonization among the Orang Asli (especially Negrito) may have arrived through recent culture integration.
It is worth noting that two separate trips were made to visit the same Kensiu village of the Negrito tribe a year apart. Two subjects whose serums were sero-negative during the first sampling were tested positive for the second sample. This indicated active transmission of H. pylori in the village. Co-incidentally, both subjects were 23 and 25 years of age. Despite the small sample size to be statistically significant, this finding serves as a warning that H. pylori colonization may be on the rise in this village.
The limitations of the present study were the relatively small number of respondents. This is due to logistic difficulties of communication, transportation and access to these Orang Asli villages. However, all possible efforts have been made in order to get as many household as possible to be interviewed. This study is also based on the self-reported prevalence which may cause report bias.

Conclusion
This study revealed that the sero-prevalence of H. pylori colonization in Malaysian Orang Asli tribes (especially Negrito) was higher than previous report. Moreover, a significant relationship was found between H. pylori sero-positivity and age, gender, tribe and building material. CagA sero-positivity was significantly associated with age, gender and tribe.