Polioviruses comprise 3 serotypes that multiply in the human intestine; in a minority of infected persons these can cause paralytic disease, poliomyelitis (1). Since the World Health Assembly adopted a resolution for the worldwide eradication of polio in 1988, the number of polio cases worldwide has declined by nearly 99.9%. Both wild poliovirus type 2 (WPV2) and WPV3 have officially been certified as globally eradicated (no cases of WPV2 have been documented since 1999, and none of WPV3 since 2012). Only two countries (Afghanistan and Pakistan) are currently affected by WPV1 (2, 3). The remarkable progress toward polio eradication is attributed to successful childhood vaccination with the oral polio vaccine (OPV) and the injectable inactivated polio vaccine (IPV) (4).

Immunization with OPV is simple and practical; it generates both systemic humoral (serum antibodies) and intestinal (secretory IgA) immunity. Vaccination with OPV induces superior intestinal immunity than IPV; thus, it can prevent the transmission of wild type viruses. Despite these advantages, OPV poses a risk, albeit very low, of vaccine-associated paralytic poliomyelitis (VAPP) (4). IPV is highly safe and induces high titers of serum antibodies (4). After the certification of WPV2 eradication, a global switch from trivalent OPV (tOPV, containing vaccine virus types 1, 2, and 3) to bivalent OPV (bOPV, containing types 1 and 3) was completed in 2016 (3, 4). At least two doses of IPV in routine immunization are also recommended in countries that vaccinate with OPV only (4, 5) to reduce the risk of VAPP. Presently, polio vaccination is based mainly on OPV (with one dose of IPV) in most low-middle income countries, while IPV is mainly used in high-income countries (4).

The threat remains of poliomyelitis re-emerging from the final global foci to spread elsewhere. This is evidenced by the silent circulation of WPV1 in 2013–2014 in the southern region of Israel, which was identified through environmental surveillance. No clinical cases of paralytic poliomyelitis were detected and molecular analysis found the virus to be similar to viruses circulating in Pakistan and from sewage samples in Egypt (6, 7). This alarming event occurred on the background of using only IPV vaccination during 2005–2013, and consistently high national coverage of polio vaccines in Israel. The silent poliovirus reintroduction, which lasted over 1 year in Israel (2013–2014), was successfully interrupted consequent to major interventions implemented by the Israeli Ministry of Health. These included mass administration of bOPV to stimulate intestinal immunity in all birth cohorts of children under age 9 years, the birth cohorts that were subject to the IPV-only vaccination program implemented during 2005–2013. In addition, OPV vaccination supplementary to IPV was reintroduced at ages 6 and 18 months, as the updated polio immunization policy in Israel (6, 8).

Oral enteric vaccines have demonstrated lower immunogenicity and efficacy in persons living in low-middle income countries, in contrast to persons from high-income countries (9). This was first reported in the 1960s and 1970s, in publications that showed OPV to be markedly less immunogenic in infants in India and other developing countries (10–13). This phenomenon was partly explained by the presence of the immunodominant attenuated type 2 poliovirus in the vaccine, which negatively affected the immune response to types 1 and 3 poliovirus in the vaccine, even as it elicited strong type 2 seroconversions (13); and to low vaccine take due to concurrent enteric infections at the time of vaccination (11, 14). An association was found of diminished immunogenicity and belonging to low socioeconomic status communities (14). Moreover, significantly lower seroconversion rates were found for poliovirus types 1, 2, and 3 following immunization with IPV in infants from Puerto Rico, compared to US infants (15). This suggested differences between the two populations, in the immune response to injectable poliovirus vaccines, as well.

Helicobacter pylori, a bacterium that persistently colonizes the stomach, is acquired in early childhood and causes asymptomatic gastritis; in a minority of infected persons peptic ulcers develop (16), and gastric cancer usually in later adulthood (17). Cytotoxin-associated gene A (CagA) antigen is main virulence factor of H. pylori. The cag pathogenicity island of H. pylori encodes for a type-IV secretory apparatus through which CagA antigen is inserted into the host cell [reviewed by Surbaum and Michetti (18)]. Infection with H. pylori CagA positive strains was linked with increased risk for peptic ulcers, premalignant gastric lesions and gastric cancer (17, 19). H. pylori have additional antigens such as VacA, Omp and NapA and others, however only a few of these antigens showed positive associations with gastric cancer (20–22). Additionally, the association between CagA sero-positivity and gastric cancer was of greater magnitude than other antigens (20, 21). Following adjustment for the presence of other antigens, CagA remained the only antigen associated with an increased risk of gastric cancer (21).

We previously showed that infection with H. pylori might affect immune responses to live oral enteric vaccines, such as Vibrio cholerae vaccine CVD 103-HgR (23) and Salmonella Typhi vaccine CVD 908-htrA (24). Specifically, the immune response was diminished in young Chilean children vaccinated with CVD 103-HgR (23), and enhanced in Malian adults vaccinated with CVD 103-HgR (25), and in US adults vaccinated with CVD 908-htrA (24). Given these discrepancies and on the background of the 2013–2014 silent outbreak with WPV1 in Israel during 2013–2014, the aim of the current study was to examine the association of H. pylori seroprevalence, and serum pepsinogens (PGs, as markers of gastric inflammation) with the neutralizing antibodies to polio vaccine strains in school-age children.

An inverse association was found between H. pylori seroprevalence accompanied with gastric inflammation, as evident by the PGI:PGII ratio, and between high neutralizing to poliovirus type 1 vaccine strain among school age children. This association was consistent and robust while using various analytical approaches to account for confounders. The association between overall H. pylori sero-prevalence and high immunity to poliovirus type 1 vaccine strain was not significant (p = 0.062).

Overall, the study population had very high immunity levels to polioviruses. This finding is in agreement with previous seroepidemiological studies from Israel (8, 40) and represents the high vaccination coverage (~95%) of poliovirus vaccines in Israel (8).

The observed inverse association between H. pylori seroprevalence/PGI:PGII ratio and high immunity to polioviruses corroborates our previous study in Chile. There we reported an inverse association between H. pylori sero-prevalence and the immune response to the live attenuated oral cholera vaccine CVD 103-HgR among children aged 6 months to 4 years. However, in older children aged 5–9 years in the same study (23), we found a positive association, as well as in Malian adults (25). Moreover, we found a positive association between H. pylori seroprevalence and the humoral immune response to S. Typhi vaccine CVD 908-htrA in US adults (24). Notably, the immune response to poliovirus in the current study results from vaccination with both intramuscular IPV in early childhood and OPV at school age, during the silent circulation of the wild type poliovirus 1. This limits direct comparison between the observations presented herein and those of previous studies on oral enteric vaccines (23, 24). Collectively, our observations suggest that H. pylori seropositivity is related to the immune response to enteric vaccines, regardless of the administration route. Moreover, our findings might have implications to children receiving combined OPV/IPV vaccination schedule, as it is the case in Israel, suggesting that the immune response elicited by combined OPV/IPV vaccines, although in general is high, might be somewhat blunted by H. pylori-associated gastric inflammation. The apparently contradictory observations support the notion that the association of H. pylori seroprevalence with the immune response to enteric vaccines might vary according to age, i.e., between children and adults. This might be explained by age-related differences in H. pylori-induced gastritis. H. pylori infection is acquired in early childhood (41), and causes chronic gastritis (42). Although most infected persons remain asymptomatic, the severity of gastric inflammation increases with age, including histopathological changes, even in the absence of symptoms (43). Antrum-predominant-H. pylori gastritis, which typically occurs in children, might enhance the secretion of gastric acid. This may explain the inverse association between the infection and an enhanced immune response to poliovirus, since the response is sensitive to acid (44, 45) if the impact is via OPV. Corpus-predominant gastritis causes hypochlorhydria, which usually occurs in adults (42). This might explain the enhanced immune response to the S. Typhi and cholera vaccines in adults (24, 25), to acid sensitive pathogens (46, 47).

In the current study, being H. pylori sero-positive and having gastric inflammation as measured by PGI:PGII <6.5 was significantly related to high immunity to poliovirus type 1 vaccine strain. However, the association with overall H. pylori sero- positivity was not significant (p=0.062), possibly due to limited power. These findings also might suggest that H. pylori might plays a role in the immune response to poliovirus vaccine only in a portion of the infected children.

Notably, lower immune response to OPV, measured in terms of seroconversion, in populations from developing countries, was studied in the 1960s and 1970s, especially the immune response to the trivalent OPV (12). This phenomenon was partly attributed to low vaccine uptake due to simultaneous enteric infections at the time of vaccination (11, 14), and the presence of the immunodominant attenuated type 2 poliovirus, which interfered with the response to types 1 and 3 in the vaccine (13). A study conducted by Swartz et al. (14), during 1969–1970 in Tel Aviv metropolitan, showed lower immune response to the trivalent OPV among infants who were vaccinated during the summer than the winter, among infants with vs. without concurrent enteric infections at the time of immunization, and among infants from low vs. high socioeconomic status (14).

H. pylori infection affects the microbiome of the stomach and gut (48, 49). A recent study showed a significant effect of enteric infections on the immune response to OPV in children from India, while the gut microbiome did not have any impact in this population (50). In contrast, another study, conducted in the USA, showed that the gut microbiome might affect the immune response to both oral and parental polio vaccines (51, 52). These studies did not assess the role of H. pylori infection. The gut microbiome might be a possible link between H. pylori infection and the adaptive immune response to vaccines. Among infants, the immune response to IPV might be reduced in relation to maternal poliovirus antibody levels (53). In adults and adolescents, increased age was associated with lower levels of polioviruses sero-positivity in Italy (54). In our study, older children had lower likelihood of having high immunity to poliovirus type 1 and type 3 vaccine strains.

Our study has several strengths. First, we used various markers to characterize H. pylori infection, including the measurement of antibodies against CagA (a virulence attribute) and serum PGs as markers of gastric inflammation. Second, we assessed several potential confounders. Third, we utilized various analytical approaches to control for confounders, all of which yielded comparable results. Lastly, the study population sample had near universal immunity to polioviruses. This enabled an exceptional opportunity to assess the relation between chronic infection and attaining the high population level of immunity to polio that is needed to prevent the re-emergence of wild type polioviruses.

Our study has some limitations, including the inability to distinguish between immunity resulting from OPV vs. IPV and between antibodies stimulated by polio vaccines vs. by clinically asymptomatic infection with wild poliovirus. Moreover, the study sample comprised a high proportion of Arab children (70% of the sample), which is not representative of the general population in Israel. However, the sample was representative of the population in the northern region of Israel, where vaccination coverage is higher than in the rest of the country. This, however, is not expected to affect our findings of inverse associations of H. pylori seroprevalence plus gastric inflammation and high-level immunity to poliovirus. These associations are generalizable to other populations.

In conclusion, H. pylori seroprevalence, accompanied by gastric inflammation, evidenced by a low PGI:PGII ratio, was inversely related to the presence of high titers of serum neutralizing antibody to poliovirus vaccines, in school age children from a population with a high background prevalence of antibodies to poliovirus.

The datasets presented in this article are not readily available because ethical and legal restrictions apply. Requests to access the datasets should be directed to Khitam Muhsen, kmuhsen@tauex.tau.ac.il.

The studies involving human participants were reviewed and approved by the Institutional Review Boards of the Hillel Yaffe Medical Center and the French Hospital, and the Ethics Committee of Tel Aviv University approved the study protocol. The parents signed a written informed consent. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

KM and ML contributed equally to the study as co-PIs and senior authors on this manuscript, planned the study, made substantial contribution to it design, acquired funding, and drafted the manuscript. EK, LB, NE, and MW directed its implementation. EK, LB, NE, KM, and ML contributed substantially to acquisition of data and blood samples. LB and MW were responsible for the laboratory work. KM, LB, and ML were responsible for data analysis. All authors made substantial contributions to interpretation of findings, approved the submitted version of the manuscript, agreed to be personally accountable for the author's own contributions and accuracy of data presented in the manuscript, and contributed to writing and substantive revisions.

This study was funded by the United States-Israel Binational Science Foundation (BSF) (PIs KM and ML) grant number 2015361.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.