The purpose of this pre-licensure cohort study was to estimate the incidence of adverse events of special interest (AESI), other adverse events (AE) leading to hospitalisation or death, meningitis and malaria in sub-Saharan African children under 5 years of age. The outcomes of this study provide the baseline data for the post-licensure EPI-MALARIA-003 (115056) study that evaluated the safety, effectiveness and impact of the RTS,S/AS01E vaccine.

An interim analysis was performed on a sub-group of study participants enrolled in active surveillance from sites where the vaccine was implemented, having 6 months of follow-up after the administration of dose 3 of DTP/HepB/Hib vaccine (6-12 weeks group), or 6 months after Visit 3 (mimicking the RTS,S/AS01E primary vaccination schedule) for the 5-17 months group; corresponding to Visit 5. The interim analysis concerned primary safety endpoints and the main secondary endpoints.

The outcomes of interest (AESI, other AE leading to hospitalisation or death, and meningitis) as well as data on malaria morbidity and mortality were captured through active and enhanced hospitalisation surveillance in children less than (<) 5 years of age. Children enrolled in active surveillance were visited at their home at regular intervals during a period of 44 months. In addition to the home visits, all visits and hospitalisations to health care facilities were recorded during this period. Thereafter there was continuous monitoring of hospitalisations only; up to study end or up to the day the child reached 5 years of age, whichever occurred first. For all children enrolled in enhanced hospitalisation surveillance, all hospitalisation events were captured during the total study period. The protocol summary has been updated with protocol amendment 4 (dated 11-Dec-2015) information:

Rationale for amendment 3:

• Participant recruitment has been modified: approximately 20,000 children were to be enrolled in the active surveillance in the 6-12 weeks group (identified at first administration of DTP/HepB/Hib vaccine), to mimic administration of RTS,S/AS01E in the 6-12 weeks age group, and 20,000 children in the 5-17 months group (either selected children identified at first administration of DTP/HepB/Hib vaccine, or children aged 5 to <18 months and corresponding to a catch-up), to mimic administration of RTS,S/AS01E in the 5-17 months age group.
• The term "passive surveillance" has been replaced by the term "enhanced hospitalisation surveillance".
• The case definitions, ascertainment and laboratory investigation (including to determine aetiology) for meningitis cases have been clarified. The objectives and outcomes related to meningitis have been modified accordingly.
• Case ascertainment for AESI has been clarified.
• It was clarified that blood sampling was to be performed as per study procedures in case of AESI or meningitis. In case of neurological AESI or meningitis, if a cerebrospinal fluid (CSF) sample was taken as part of routine practice, part of the sample was stored. Only serious adverse events related to blood sampling were collected.
• Limitations due to the sample size and the study design have been further discussed.
• For clarity, consistency and comparability in the objectives, outcomes and methodologies, protocol amendment 3 text has been aligned in EPI-MALARIA-002 (115055) with the EPI-MALARIA-003 (115056) protocol.

Rationale for amendment 4:

• The background section has been updated with data about cerebral malaria from a post-hoc analysis of the Phase III study MALARIA-055.
• Additional secondary objectives have been added for estimation of probable meningitis and for estimation of the incidence of cerebral malaria using RDT and/or microscopy.
• The design and the analysis have been modified to take into account implementation of a 4th dose of RTS,S/AS01E. In addition, the follow-up period after the last dose has been extended to correspond to the follow-up period of the children enrolled in EPI-MAL-003 (i.e. 24 months after the 4th dose of RTS,S/AS01E). Indeed, home visits at 12 months and 24 months after the last RTS,S/AS01E dose will help to capture protocol-defined diseases that may have long risk window periods and that may not have been identified if the subject did not visit a health care facility, and to monitor the occurrence of malaria episodes for evaluation of vaccine effect. The age of the study population has been adapted accordingly (< 5 years).
• The section about sites participating to the study was updated following SAGE/MPAC recommendations of pilot implementations of RTS,S/AS01E in 3-5 distinct settings in SSA restricted to moderate-to-high transmission of malaria.
• The activities related to home visits to detect malaria cases, including training of community health workers for systematic measurement of body temperature of all children, and availability of malaria tests have been clarified for alignment with the EPI-MAL-003 protocol.
• It has been clarified that cases of malaria only detected during the annual visits planned for EPI-MAL-005 will not be included in the analysis of EPI-MAL-002. However, if the cases of malaria are detected during an EPI-MAL-005 home visit that coincides with a home visit scheduled in EPI-MAL-002, the events will be captured in EPI-MAL-002.
• The case definitions for malaria have been revised, according to the 3rd edition of the WHO guidelines for the treatment of malaria (2015). In addition, a case definition for cerebral malaria has been added.
• Case ascertainment by the external panel of experts has been clarified.
• It has been clarified that severe/cerebral malaria cases and other AE leading to hospitalisation or death were to be reviewed by the external panel of experts.
• The incidence of other AE leading to hospitalisation or death, meningitis and malaria morbidity and mortality were monitored in sub-populations of children with hemoglobinopathies and HIV-positive children.
• The section about analysis of co-primary objectives has been revised to mention that additional at risk periods were considered based on results as sensitivity analyses.
• The section about analysis of secondary objectives has been revised for clarification of potential variables that may be considered for models, in line with information collected in eCRF, and for estimation of the incidence of cerebral malaria.
• It has been clarified that an interim analysis was to be performed with the data collected on all subjects after 6 months of follow-up following the administration of dose 3 of DTP/HepB/Hib vaccine (6-12 weeks group), or 6 months after V3 (5-17 months group).
• The section about handling of missing data has been revised.
• It has been clarified that data about seasonal malaria chemoprevention was to be collected.

Other changes were made for simplification, clarification or consistency.

Rationale for amendment 5:

• In order to have a synergy with the WHO pilot implementation, the study size was reduced from 40,000 to 30,000 children in active surveillance with at least 20,000 children enrolled where the RTS,S/AS01E vaccine was implemented. Among the 30,000 children, approximately 15,000 children (with at least 10,000 children in sites where the vaccine was implemented) were to be enrolled in the 6-12 weeks group (to collect background data in this age group) and approximately 15,000 children (with at least 10,000 children in sites where the vaccine was implemented) were to be enrolled in the 5-17 months group (to mimic administration of RTS,S/AS01E in the 5-17 months age group). Study size updates involved multiple sections, e.g. Section 9.1 Study Design and Section 9.5 Study size.
• For multiple sections: the number of countries and number of study sites has been updated following the WHO announcement that the RTS,S/AS01E vaccine was first introduced in 3 countries (Ghana, Kenya and Malawi) through the MVIP.
• For multiple sections: estimation of the mortality rate, overall and by gender, has been added.
• For multiple sections: the informed consent process has been clarified "signed or witnessed and thumbprinted ICF".
• For multiple sections: "or equivalent surveillance system" has been added after Health and Demographic Surveillance System (HDSS) and the term "study site(s)" or "study area" has been added.
• For multiple sections: the replacement strategy has been updated.
• For multiple sections: the duration of the recruitment period has been removed, as it could/cannot be respected in all sites due to logistical constraints.
• For multiple sections: it was clarified that census rounds are scheduled at least once a year during the study periods instead of twice a year.
• The milestones in Section 6 have been updated.
• New sub-sections on "Severe malaria including cerebral malaria" and "Overall mortality by gender" have been added to Section 7.2 Safety results of the clinical development of RTS,S/AS01E.
• The wording of "secondary objectives" on "describe the causes of hospitalisation" and "estimate the incidence of cerebral malaria" has been changed, and the secondary objective "To describe the causes of death, overall and by gender" has been added in Section 8.2 and in the corresponding part of Section 4 Abstract.
• The opening paragraph of Section 8.2 Secondary Objectives and of the corresponding part of Section 4 Abstract has been updated.
• The wording of the last two "study variables" / "study endpoints" on "Occurrence of hospitalisations" and "Occurrence of death" has been modified in Section 9.3.2 and in the corresponding part of Section 4 Abstract.
• In Section 9.2.1 Study population: the text on "Estimated annual birth cohorts by study site" (Table 2) has been removed, with removal of planned number of sites in selected countries.
• Section 9.2.5.3 Meningitis has been edited, including addition of a new opening paragraph, addition of "turbid macroscopic aspect" as a sign of CSF abnormality, and clarification that clinically suspected meningitis cases included subjects with no CSF sample available and no alternative diagnosis.
• For consistency between the way the objectives and the endpoints were presented, and to avoid repeating several endpoints, the two sub-sections in Section 9.3.2 Secondary endpoints: "In children included in active or enhanced hospitalisation surveillance, prior to implementation of RTS,S/AS01E" and "In children included in active surveillance, prior to implementation of RTS,S/AS01E" have been replaced by one sub-section: "In children living in the study area, prior to implementation of RTS,S/AS01E", followed by all the secondary endpoints.
• In Section 9.4.3 Study EPI-MAL-005 "changes in environmental factors such as rainfall" has been added to the parameters assessed in EPI-MAL-005.
• Section 9.5 Study size has been updated as shown in bullet 1 above with greater detail on the estimates of observed incidence of events following dose administration based on either: 15,000 subjects, 10,000 subjects, the estimated birth cohort of 11,500 subjects and an estimated birth cohort of 17,250 children in Tables 6-9.
• Section 9.7.6 Analysis of secondary safety objectives has been updated, including analysis of cerebral malaria of the causes of death, overall and by gender.

Finally, changes in study personnel have been included on the protocol cover page

Rationale for amendment 6 Protocol Amendment 6 has been put in place to document that the EPI-MAL-002 study sites in Burkina Faso early terminated the study and that sites in Malawi, one of the countries where RTS,S/AS01E was introduced through the MVIP will not take part in the study.

GSK, in agreement with WHO, decided to terminate the study EPI-MAL-002 in the two sites located in Burkina Faso, i.e. Nouna and Sapone. Briefly, it was previously decided that Burkina Faso sites would continue participating in EPI-MAL-002 until study end. However, given that the MVIP will not be conducted in this country, there is no scientific value to collect additional data (the EPI-MAL-002 data collected in Burkina Faso will neither be used for the before-after analyses, nor for the generation of EPI-MAL-002 indicators that inform any other analyses of the EPI-MAL-003 study). Therefore, the study EPI-MAL-002 in Burkina Faso sites has been early terminated and data collected and recorded to date in those sites was to be reported in a descriptive way.

In order to align with the MVIP, the study sites for the GSK Phase IV studies have been selected from the 3 countries where the RTS,S/AS01E vaccine was implemented (Ghana, Kenya and Malawi). Considering the RTS,S/AS01E vaccine implementation date in Malawi was planned in October 2018, the baseline data that might be collected in Malawi in the EPI-MAL-002 study would be too limited to be relevant for the before/after comparisons in this country. Therefore, GSK, in agreement with the WHO, decided to focus the conduct of the EPI-MAL-002 study in Ghana and Kenya, not initiating the study in Malawi, and partially compensating the expected sample size from Malawi sites by using the high recruitment observed from sites in Kenya (Kombewa) and Ghana (Kintampo) and extending recruitment in Ghana (Navrongo), hence limiting the impact on the before/after comparison study power.

Inclusion Criteria:

All subjects must satisfy ALL the following criteria at study entry:

• Subjects' parent(s)/Legally Acceptable Representative(s) [LAR(s)] who, in the opinion of the investigator, can and will comply with the requirements of the protocol.
• Written informed consent provided from either the parent(s) or LAR of the subject.
• Subject living in the Health and Demographic Surveillance System (HDSS) area.
• For enrolment in active surveillance: children must be <18 months of age. OR
• For enrolment in enhanced hospitalisation surveillance: children must be <5 years of age and hospitalised at any time during the study.

Exclusion Criteria:

• Child in care.

The study population was defined as those children living in the study areas who are < 5 years old.

GSK agreements may vary with individual investigators, but will not prohibit any investigator from publishing. GSK supports the publication of results from all centers of a multi-center trial but requests that reports based on single site data not precede the primary publication of the entire clinic.