Increased bacterial taxonomic and functional diversity is associated with impaired rotavirus vaccine immunogenicity in infants from India and Malawi

Table 1 Baseline characteristics of study cohorts

	India	Malawi
N	283	58
Caesarean delivery	62/283 (21.9)	0 (0.0)
Birthweight (kg), mean (s.d.)	2.96 (0.43)	3.02 (0.45)
Female	152 (49.5)	60 (50.4)
Polio vaccine schedule
Trivalent OPV	51 (18.0)	–
Mixed trivalent/bivalent OPV^a	56 (19.8)	–
Bivalent OPV^a	84 (29.7)	58 (100.0)
IPV^b	92 (32.5)	–
Exclusive breastfeeding up to 11 weeks of life	243 (85.9)	53 (91.4)
Exposed to antibiotics up to 14 weeks of life	204 (72.1)	43 (74.1)
HIV status
Exposed	0 (0.0)	15 (25.9)
Unexposed	275 (97.2)	41 (70.7)
Unknown	8 (2.8)	0 (0.0)
Neonatal rotavirus infection^c	149/281^d (53.0)	6/47^d (12.8)
Seroconverted to ORV	78/282^d (27.7)	10/56^d (17.9)
Shed ORV at week of life 7 (1 week after dose 1)	74/281^d (26.3)	21/47^d (44.7)
RV-IgA at week of life 6 (before dose 1), GM (95% CI)	7.3 (5.9–9.1)^e	3.4 (2.5–4.6)
RV-IgA at week of life 14 (4 weeks after dose 2), GM (95% CI)	18.9 (14.9–23.9)	6.4 (4.0–10.2)

Data are n (%) unless otherwise stated. Seroconversion was defined as detection of RV-IgA at ≥ 20 IU/ml post-vaccination among infants who were seronegative at baseline or a fourfold increase in RV-IgA concentration among infants who were seropositive at baseline. Neonatal rotavirus infection was defined as the detection of wild-type rotavirus shedding in week of life 1 and/or baseline seropositivity (pre-vaccination RV-IgA ≥ 20 IU/ml)
GM geometric mean, IPV inactivated poliovirus vaccine, OPV oral poliovirus vaccine, ORV oral rotavirus vaccine, RV rotavirus
^aEnrolment spanned the global transition from trivalent to bivalent OPV in April 2016
^bA sequentially recruited cohort in India received IPV instead of OPV at 6 and 10 weeks of age
^cAmong 149 Indian infants with neonatal infection, 120 (80.5%) were positive for wild-type rotavirus shedding in week of life 1, 90 were seropositive for RV-IgA at week of life 6 (60.4%), with an overlap of 61 (40.9%) positive via both methods
^dInfants with missing data excluded
^dAmong neonatally infected infants, RV-IgA GM (95% CI) was 23.8 (18.0–31.7) before vaccination

ISSN: 1471-2180