Seroprevalence of poliovirus antibodies in Nigeria: refining strategies to sustain the eradication effort

Introduction in 2016, a switch from trivalent oral poliovirus vaccine (OPV) (containing serotypes 1,2,3) to bivalent OPV (types 1,3) was implemented globally. We assessed the seroprevalence of poliovirus antibody levels in selected Nigerian states, before and after the switch, documented poliovirus type2 outbreak responses conducted and ascertained factors associated with immunity gaps based on seroprevalence rates. Methods we conducted a secondary analysis of stored serum samples from the 2018 Nigeria National HIV/AIDS Indicator and Impact Survey. Serum from 1,185 children aged 0-119 months residing in one southern and four northern states were tested for serotype-specific PV neutralizing antibodies; seropositivity was a reciprocal titer ≥8. We conducted regression analysis to determine sociodemographic risk factors associated with low seroprevalence using SAS 9.4. Results children aged 24-119 months (pre-switch cohort) had seroprevalence against PV1, PV2, and PV3, of 97.3% (95% CI:96.4-98.2), 93.8% (95% CI:92.2-95.5), and 91.3% (95% CI:89.2-93.4), while children aged <24 months (post-switch) had seroprevalence of 86.0% (95% CI:81.2-90.8), 55.6% (95% CI: 47.7-63.4), and 77.2% (95% CI:71.0-83.4) respectively. Regression analysis showed age <24 months was associated with lower seroprevalence against all PV serotypes, (p<0.0001); females had lower seroprevalence against PV1 (p=0.0184) and PV2 (p=0.0354); northern states lower seroprevalence against PV1 (p=0.0039), while well-water source lower seroprevalence against PV3 (p=0.0288). Conclusion this study showed high seroprevalence rates against PV 1, 2, and 3 in pre-switch children (aged 24-119 months). However, post-switch children (<24 months) had low immunity against PV2 despite outbreak responses. Strategies to increase routine immunization coverage and high-quality polio campaigns can increase immunity against polio virus.


Introduction
During the 41st World Health Assembly in 1988, World Health Organization (WHO) Member States committed to the global eradication of poliomyelitis [1].Oral poliovirus vaccine (OPV) is a highly effective public health intervention deployed to achieve polio eradication and has reduced the burden of polio by more than 99.99% globally since 1988.Wild poliovirus (WPV) type 2 and WPV type 3 were certified eradicated globally in 2015 and 2019, respectively [2,3].Afghanistan and Pakistan remain the only two countries endemic with WPV type 1 after the certification of countries of the WHO African Region, including Nigeria, as WPV-free in August 2020 [4].
The live, attenuated Sabin OPV used for vaccination, while undergoing prolonged transmission in under-immunized populations, can genetically revert to a virulent strain called vaccine-derived poliovirus (VDPV).Vaccine-derived polioviruses are biologically equivalent to WPVs and can cause paralytic poliomyelitis among under-immunized or nonimmunized people.When transmission of VDPVs is confirmed, it is deemed an outbreak and the virus is denoted as a circulating (cVDPV).To prevent outbreaks of cVDPV type 2 (cVDPV2) emerging in areas with low polio immunity, the Global Polio Eradication Initiative (GPEI) implemented the global withdrawal of OPV type 2 for routine vaccination following certification of WPV type 2 eradication.
In April 2016, as an important aspect of OPV type 2 withdrawal, GPEI launched the globally synchronized switch from trivalent OPV (tOPV, containing Sabin strain types 1, 2, and 3) to the bivalent OPV (bOPV, containing only Sabin strain types 1 and 3).One dose of injectable inactivated poliovirus vaccine (IPV, containing types 1, 2 and 3) was introduced into routine immunization (RI) schedules to maintain some level of poliovirus immunity among children against all 3 poliovirus types [5] and reduce the risk of cVDPV2 paralytic disease.
The level of immunity against polio types 1, 2, and 3 from seroprevalence surveys in different states in Nigeria and time points before the interruption of WPV transmission revealed significant variability in population immunity [9][10][11][12].We determined the seroprevalence of neutralizing poliovirus antibody levels in stored serum specimens collected from children during a nationwide 2018 Nigeria HIV serosurvey, following the withdrawal of tOPV and with the implementation of supplementary immunization activities (SIAs) with mOPV2 in cVDPV2 outbreak responses and assessed the factors associated with low seroprevalence.Findings on immunity gaps from this study will inform the development of evidencebased immunization strategies for outbreak responses and routine immunization intensification.

Methods
Study setting, design and population: this was a secondary analysis of stored samples of the 2018 Nigeria National HIV/AIDS Indicator and Impact Survey (NAIIS).National HIV/AIDS Indicator and Impact Survey was a nationally representative, population-based, cross-sectional household survey conducted in all 36 states and the Federal Capital Territory (FCT) from July to December 2018 to determine national and subnational HIV incidence, prevalence, and viral load.Several variables were included in the data collection instruments; however, immunization histories were not collected.A total of 205,903 participants, including the caretakers of 32,354 children (aged 0-14 years) and 173,549 adults (aged 15-64 years), consented to storage and future testing of specimens for diseases of public health importance.This study and analysis focused on children aged 0 to 119 months (i.e.<10 years).The NAIIS used a two-stage cluster sampling technique, selecting enumeration areas followed by households and then participants.The details of sampling for the NAIIS survey have been reported [13].Infants born after tOPV removal from the country´s immunization program in April 2016 to the time of sample collection of the NAIIS 2018 are defined as the postswitch population.Due to variability by state from the time of the tOPV switch to sample collection for NAIIS, post-switch children were restricted to children aged <24 months to reduce the risk of miscategorization of the post-switch population; pre-switch children were those aged 24-119 months included a fraction of post-switch children in the group aged 24-35 months.
Blood samples collection and storage: data and blood samples during NAIIS were collected from June-December 2018.Over 98% of participants in the NAIIS study granted consent for leftover samples to be stored and tested for other diseases of public health importance.Residual sera were stored at -80°C.Stored samples from four northern states (Borno, Jigawa, Yobo, Sokoto) and one southern state (Lagos) were selected as described below.Selected sera were shipped to the Population Immunity Laboratory at the Centers for Disease Control and Prevention in Atlanta, Georgia, USA.Samples were tested for the presence of neutralizing antibodies against poliovirus types 1, 2, and 3 using a standard microneutralization assay [14].Seropositivity was defined as reciprocal titers of poliovirus neutralizing antibody ≥1:8.
Sample size: sample size for each state was determined based on the expected poliovirus seroprevalence using data from similar studies [9,10,12,15] and the target confidence interval (CI) adjusted for design effect.We assumed 60% overall seroprevalence in the northern states, 80% in Lagos State (a southern state), a target CI of 0.14, and a design effect of 2. The expected sample size was 1,224 for northern states (408 per state) with ~150 samples for children aged <24 months and 258 for 24-119 months per state; the sample size for Lagos State was 279.There were approximately 93 samples for each of the five age group strata.The total estimated sample size was 1,508.Due to the variability in poliovirus seroprevalence by age and the likelihood of lower seroprevalence among children aged <24 months, specimens for all children in this age group with consent for additional testing of stored specimens were included in this study; children aged 24-119 months were selected via simple random sampling and tested based on available specimens.
Statistical analysis: we estimated the overall and state-level seroprevalence, expressed as percentages with 95% CIs, against poliovirus types 1, 2, and 3 by age strata and by socio-demographic groups: sex, region, type of residence, water source, toilet type, and wealth index.The NAIIS 2018 wealth Index was constructed using household characteristics, asset ownership variables, services, and amenities that were present in both urban and rural surveyed areas, assumed to be good indicators of economic status [16].The wealth index was presented as low, middle, or high income [13].The distribution of poliovirus neutralization titers is presented using the median log2 and 95%CI for each of the three serotypes.Specimens without detectable neutralizing antibodies (<3.00 log2) are given a value of 2.50 log2.We conducted logistic regression analyses to assess factors associated with low seroprevalence including sex, age, region, toilet type, and water source.We documented the number of mOPV2 polio campaigns held during 2016-2018 [17] before NAIIS and compared it with the seroprevalence rates for children aged <24 months by state.All data analyses were performed in SAS 9.4 (SAS Institute Cary NC, USA).
Ethical consideration: the study proposal for testing of residual samples after NAISS HIV testing received approvals from the Nigeria Health Research and Ethics Committee (NHRECC) and the University of Maryland Baltimore Institutional Review Board.The US CDC approved the project as a non-research public health surveillance activity.The participant´s parents or guardians gave written informed consent and permission for residual samples to be stored and tested for diseases of public health importance.

Results
In total, 1,185 samples were tested for types 1, 2, and 3 poliovirusneutralizing antibodies, including 238 samples from children aged <24 months.Half (n=597, 50.4%) of the children were male, 49.4% were aged ≥ 60 months, and 50.8% were living in rural areas.The main water sources for participants were pipe-borne water (63%).Only 25% had access to a flush toilet system.Most of the children (61.6%) were in the low-wealth index category (Table 1).
Overall, the seroprevalence of poliovirus-neutralizing antibodies was 95.7% (95% CI: 94.7-96.7)for type 1,88.6%(95% CI: 86.8-90.4) for type 2, and 89.3% (95% CI: 87.4-91.3)for types 3. Seroprevalence for type 1 and type 3 polio antibodies was higher in the southern region than northern region.Among all the states in the northern region, Sokoto State had the lowest seroprevalence rates (Table 2).Seroprevalence was higher among those residing in urban areas compared to those residing in rural areas for all three polio types.By water source, those with piped water, bottle, or borehole had higher seroprevalence than those with well or spring water for all three types.By toilet type, those with a flushed system had slightly higher seroprevalence against poliovirus type 1 (97.5, 95% CI: 95.9-99.1)than those with pit latrine (94.4,95% CI: 92.8-95.9)and those with no facility/bush (95.2, 95% CI: 92.9-97.5).Those children from high-wealth-index families had higher seroprevalence for all three types than those from low-wealth-index families.The seroprevalence generally increased with age (Figure 1).The age groups of 0-11 months and 12-23 months had the lowest seroprevalence for types 1, 2, and 3 compared to older age groups (Table 2).Children <24 months had the lowest PV2 seroprevalence rate of 55% (57% for children 0-11 months and 54% for 12-23 months of age (Table 2) irrespective of the number of mOPV2 campaigns conducted in the same year prior to NAIIS.
In 2018, all northern states had carried out two to four rounds of mOPV2 campaigns and one intensified IPV round 1-12 weeks prior to the NAIIS sample collection.However, seroprevalence for type 2 in children 0-23 months was low; no mOPV2 or intensified IPV campaigns were held in Lagos State (Table 5).

Discussion
This study showed overall higher seroprevalence in children aged <10 years of age for poliovirus types 1 (96%) than types 2 and 3 (88.6%and 89%, respectively).However, the post-switch population of children <24 months of age had lower seroprevalence for all serotypes, especially type 2, at a seroprevalence rate of 55%.This age group is the population most vulnerable to poliomyelitis [18].Regression analysis confirmed that with all other variables controlled, children <24 months had significantly lower type 2 immunity despite opportunities for IPV and mOPV2 vaccination.This finding is consistent with prior studies in Nigeria [9,15], but it is concerning due to the unprecedented number and geographic spread of cVPV2 outbreaks in Nigeria following the switch.
Prior to sample collection for NAIIS In 2018, at least 2 mOPV2 outbreak response SIAs were carried out in each of the included northern states, type 2 seroprevalence ranged 50-68%.Routine immunization coverage with IPV is low in the northern states, with the 2018 DHS survey results ranging from 20-45% in the northern states, compared to 87% in Lagos State [8].Low IPV coverage means many children are vulnerable to paralysis if infected with cVPV2.There were 34 cVDPV2 AFP cases reported in the northern states during 2016-2018 compared with none in Lagos State during this period [17].The indicator of the quality of the cVDPV2 outbreak response mOPV2 SIAs in these northern states was suboptimal; lot quality assurance sampling (LQAS) surveys [19] The Pan African Medical Journal.2023;45 (Supp 2):2 | Omotayo Bolu et al.
results at the local government area level ranged 65-85% passing at a 90% threshold.With low-quality SIAs, a substantial proportion of unimmunized children remains.Concordant to the LQAS survey results, type 2 seroprevalence in children aged <24 months in this study was only 55%.There were several limitations to the study.First, it was underpowered.The number of samples tested and used in analyses was less than the estimated sample size since NAIIS sampling was not intended to assess poliovirus seroprevalence; in addition, some stored samples had insufficient serum volume for testing.Nonetheless, this study had one of the larger sample sizes conducted for seroprevalence studies in Nigeria [10,12,15].Second, because of the small number of samples available from Borno and Yobe State, we combined Borno and Yobe as a cluster and there are no state-specific estimates.Third, data were not collected on the number of OPV or IPV vaccine doses that participating children received as part of the RI program or during SIAs.Thus, we could not compare seroprevalence survey data with the number of vaccines each child had received.However, we had data on mOPV2 SIAs and IPV intensification by state to indicate opportunities for type 2 seroconversion.
Despite these limitations, this study reflects low immunity in children born post-switch in the northern states which indicates low RI IPV coverage, and poor-quality mOPV2 and IPV SIAs that repeatedly did not reach target populations.As anticipated, our findings reflect that the switch from tOPV to bOPV, prior to having sufficient IPV coverage, contributed to type 2 immunity gaps.With the novel coronavirus disease 2019 (COVID-19) pandemic, when RI coverage decreased further and bOPV2 campaigns were not conducted, waning immunity against types 1 and 3 will occur as well unless bOPV and IPV administered in RI are improved in parallel with preventive bOPV SIAs reinstated and type 2 response SIAs are of higher quality.Our data indicate that a large proportion of children in the northern states were missed during mOPV2 outbreak response SIAs.Lagos State also had low type 2 immunity among post-switch children aged <24 months as no mOPV2 response SIAs were implemented during the study period.
Implementation or improvement of several strategies would help to interrupt cVDPV2 outbreaks and improve RI coverage of poliovirus vaccines.First, Nigeria could work on performing higher-quality OPV SIAs with a heightened focus on reaching populations that have been repeatedly missed during campaigns.Chronically missed children and settlements sustain cVDPV2 transmission.Therefore, careful evaluations of all settlements in the catchment area, frequent mapping using geospatial technology, strong pre-campaign preparedness and enhanced supervision of SIA teams are critical for improving campaign quality and reducing the number of repeatedly missed children [20].
Other approaches can be implemented like nomadic population enumeration, which helped identify several missed children in 2014, to ensure more children are reached with vaccines [21].To prevent seeding new cVDPV2 outbreaks with suboptimal mOPV2 SIAs, novel OPV type 2 (nOPV2) can be used for type 2 outbreak response SIAs.The nOPV2 is modified to be more genetically stable than Sabin mOPV2 and less likely to be associated with the emergence of cVDPV2 in low immunity settings [22].In clinical trials, nOPV2 provides comparable seroconversion against type 2 poliovirus [23].Following emergency use authorization by the World Health Organization (WHO), nOPV2 is now being used in Nigeria [22,24] and other countries, although supply has been limited.Secondly, while improving outbreak response SIAs, there could be a parallel effort on increasing RI coverage.Improving political, social, and cultural interventions can ensure caregivers attend RI scheduled visits, as well as take children to receive poliovirus vaccines during SIAs.Thirdly, IPV intensification can be conducted in states with low polio serotype 2 coverage, which has been effective in other settings to improve immunity for all three serotypes [22,25,26].Like OPV SIAs, which achieved higher LQAs results relative to the IPV campaigns in 2018 in Nigeria [17], similar preparedness steps can be taken to improve IPV coverage.
Since lack of access to potable water and being a girl were risk factors for lower type 2 polio seroprevalence, GPEI and local leaders could engage with ministries in charge of water and sanitation to enhance access over time.In the interim, attention could be paid to sex disparities wherein females may be more selectively missed during SIAs campaigns and have decreased access to RI services.Nigeria introduced a second dose of IPV in September 2021; the current vaccination schedule includes IPV at ages 6 weeks and 14 weeks.Every effort to get infants to receive both IPV doses can be put in place while also enhancing all efforts to ensure quality OPV campaigns.
Implication of the study findings: the findings show that type 2 immunity is low in the selected states in children aged 0-23 months even after several outbreak response campaigns with mOPV2.Increase focus on reaching children aged 0-23 months during outbreak responses with type 2 OPV campaigns, while improving routine immunization coverage with all poliovirus vaccines.

Conclusion
This study showed high seroprevalence rates against PV types 1, 2, and 3 in children aged 24-119 months.However, children born after the switch (<24 months) had low immunity against PV2 despite two or more rounds of outbreak responses using mOPV2.Age, access to clean water sources and latrines were associated with lower polio seroprevalence indicating immunity gaps.Strategies to increase routine immunization coverage, especially for post-switch children including periodic intensification of routine immunization and high-quality polio campaigns can increase immunity against the polio virus.