GACVS held its 41st meeting in Geneva, Switzerland, on 4–5 December 2019.² The Committee examined data on the safety of vaccines against rotavirus.

The occurrence and diagnosis of intussusception varies by geographical region. Age-specific incidence is characterized by a sharp increase during the first 6 months of life, which coincides with the age at which most primary vaccines, including rotavirus vaccine (RVV), are administered. Currently, 4 WHO-pre-qualified RVVs are available. The purpose of the session was to review data on the safety of RotaTeq™ in 5 sub-Saharan African countries and of RotaVac™ in parts of India and to examine baseline data from a nationwide network in India of active surveillance for intussusception following vaccination.

Experience in 5 sub-Saharan African countries

The African Intussusception Surveillance Network was formally established in 2014 to monitor the risk of intussusception with 2 vaccines introduced in the region, Rotarix™ and RotaTeq™. Use of a common protocol at all sites in all countries allowed pooling of data for analysis. GACVS previously reviewed findings related to Rotarix™.³ The 5 countries that have used RotaTeq™ are Burkina Faso, Côte d’Ivoire, Gambia, Mali and Rwanda.

A systematic review4 was conducted to update a 2012 Cochrane systematic review regarding the efficacy and safety of rotavirus vaccines. This review included RCTs (low power, low bias); historical controls (low power, high risk of bias); case-control studies (high power, high risk of bias); cohort studies (high power, high risk of bias); and self-controlled case series (SCCS) (high power, unclear risk of bias). Data were insufficient to evaluate many of the new vaccine products. For those reviews with sufficient data, evidence from RCTs showed that there was no difference in incidence of serious adverse events in the use of RV1, RV5, Rotasiil®, or Rotavac® compared with placebo, up to 2 years after vaccination. There was conflicting evidence from different sources as to whether RV1 or RV5 was associated with an increased risk of intussusception. While RCTs of RV1 and RV5 found no association between intussusception and vaccination, SCCS studies suggested an increased risk in the weeks following vaccination.

Infants with onset of symptoms of intussusception between the ages of 28 and 245 days (4 weeks to 8 months) were included between October 2012 and October 2019. The surveillance period varied by country, according to the date of introduction of the vaccine and when the country joined the network. Similarly, the end of the surveillance period was based on when it switched vaccine type or last submitted information to the database. The dates of vaccination against rotavirus were obtained from vaccine cards or medical or clinical records. There were 1–5 sites in each participating country. Altogether, 275 cases of intussusception were included in the self-controlled case series (SCCS) analysis, with 82 (30%) from Burkina Faso, 36 (13%) from Côte d’Ivoire, 2 (1%) from Gambia, 102 (37%) from Mali and 53 (19%) from Rwanda. The relative incidence of intussusception during days 1–7 was 4.11 (95% confidence interval 0.79–11.52) after dose 1, 0 after dose 2 and 0.86 (95% confidence interval 0.28–1.92) after dose 3. The non-statistically significant increase in risk of intussusception during the first 7 days after dose 1 as compared with the period ≥ 28 days is consistent with findings from other parts of the world.⁴ The analysis was conducted before the expected case sample size was reached, because RotaTeq™ will no longer be supplied to countries considered eligible by Gavi, the Vaccine Alliance (Gavi). As different products will be used in the future, continued active surveillance is desirable to monitor the product-specific risk.

Risk of intussusception with RotaVac™ in India

RotaVac™ was introduced into India’s Universal Immunization Programme in phases, beginning with 4 states in 2016. Three doses of vaccine are recommended, at 6, 10 and 14 weeks of age; the first dose can be given up to 1 year of age. Active sentinel surveillance was established in 28 hospitals in 9 states. The analysis was based on cases that met the Brighton Collaboration definition of level 1 of diagnostic certainty. Clinical features, date of symptom onset, mode of treatment, socio-demographic characteristics, an ultrasound report and image, hospital procedure and treatment notes were recorded. The vaccination status of infants was verified from vaccination cards and registers of workers in rural child care centres. SCCS analysis was conducted to determine whether there was an association between oral rotavirus vaccination and intussusception in 589 children with ascertained vaccination status (0, 1, 2 or 3 doses of RotaVac™). The incidence rate ratios for the first 7 days after doses 1, 2 and 3 as compared with the period from day 28 to 1 year of age were 0.83, 0.86 and 1.65, respectively, none of which was statistically significantly different from no association. Similarly, no significant difference was found for the 8–21-day risk window after vaccination.

Baseline data from 2010 to 2017 were also presented from an Indian nationwide network of tertiary care hospitals in 18 states. Most of the data were from a period before introduction of RVV. Data on intussusception were collected for children aged <2 years. Over 75% of cases occurred before 12 months of age and the majority during the first 6 months. The numbers of cases were age-dependent, with a peak between 20 and 35 weeks of age. The baseline hospitalization rates for intussusception (number of cases per 1000 paediatric surgical admissions) differed by region, with the highest in southern states and the lowest in western India

Conclusions and recommendation of the Committee

The incidence of intussusception varies by population and infant age. The data did not indicate a significantly higher risk of intussusception during the post-vaccination risk periods than in the reference period for either of the vaccines (RotaTeq™ and RotaVac™). Continued monitoring of the risk for intussusception associated with RVVs and comparisons of products in the same risk window are recommended when new RVVs are introduced into new populations. As common definitions and protocols have been used, India may consider pooled analysis of their intussusception data to increase analytical power.

² GACVS invited additional experts to present and discuss evidence related to particular topics. These experts were affiliated with: Crozet BioPharma LLC, Devens (MA), USA; Task Force for Global Health, Decatur (GA), USA; Paul-Ehrlich-Institut, Langen, Germany; National Institute of Allergy and Infectious Diseases, Bethesda (MD), USA; University of Jordan, Amman, Jordan; Johns Hopkins Bloomberg School of Public Health, Baltimore (MD), USA; Centers for Disease Control and Prevention, Atlanta (GA), USA; and Janssen Pharmaceutica, Bersee, Belgium.

³ See No. 3, 2018, pp. 19–21.

⁴ See No. 6, 2012, pp. 54–56.

Full report of GACVS meeting of 4-5 December 2019, published in the Weekly Epidemiological Record on 24 January 2020