The effect of live attenuated influenza vaccine on pneumococcal colonisation densities among children aged 24–59 months in The Gambia: a phase 4, open label, randomised, controlled trial

BACKGROUND: Influenza and other respiratory viruses promote Streptococcus pneumoniae proliferation in the upper respiratory tract. We sought to investigate for what we believe is the first time, the effect of intranasal live attenuated influenza vaccine (LAIV) on nasopharyngeal S pneumoniae density in a low-income to middle-income country population with high pneumococcal carriage rates. METHODS: In an open-label, randomised, controlled trial in The Gambia, 330 healthy children aged 24–59 months were randomly assigned 2:1 to receive one trivalent LAIV dose at enrolment (day 0, intervention) or at the end of active follow-up (day 21, control). The investigator team were initially masked to block size and randomisation sequence to avoid allocation bias. Group allocation was later revealed to the investigator team. The primary outcome was PCR-quantified day 7 and 21 pneumococcal density. Asymptomatic respiratory viral infection at baseline and LAIV strain shedding were included as covariates in generalised mixed-effects models, to assess the effect of LAIV and other variables on pneumococcal densities. The study is registered at ClinicalTrials.gov, NCT02972957, and is closed to recruitment. FINDINGS: Between Feb 8 and April 12, 2017, and Jan 15 and March 28, 2018, of 343 children assessed for eligibility, 213 in the intervention group and 108 in the control group completed the study and were included in the final analysis. Although no significant differences were seen in pneumococcal carriage or density at each timepoint when comparing groups, changes from baseline were observed in the LAIV group. The baseline S pneumoniae carriage prevalence was high in both LAIV and control groups (75%) and increased by day 21 in the LAIV group (85%, p=0·0037), but not in the control group (79%, p=0·44). An increase in pneumococcal density from day 0 amounts was seen in the LAIV group at day 7 (+0·207 log(10) copies per μL, SE 0·105, p=0·050) and day 21 (+0·280 log(10) copies per μL, SE 0·105, p=0·0082), but not in the control group. Older age was associated with lower pneumococcal density (−0·015 log(10) copies per μL, SE 0·005, p=0·0030), with the presence of asymptomatic respiratory viruses at baseline (+0·259 log(10) copies per μL, SE 0·097, p=0·017), and greater LAIV shedding at day 7 (+0·380 log(10) copies per μL, SE 0·167, p=0·024) associated with higher pneumococcal density. A significant increase in rhinorrhoea was reported in the LAIV group compared with the control group children during the first 7 days of the study (103 [48%] of 213, compared with 25 [23%] of 108, p<0·0001), and between day 7 and 21 (108 [51%] of 213, compared with 28 [26%] of 108, p<0·0001). INTERPRETATION: LAIV was associated with a modest increase in nasopharyngeal pneumococcal carriage and density in the 21 days following vaccination, with the increase in density lower in magnitude than previously described in the UK. This increase was accelerated when LAIV was administered in the presence of pre-existing asymptomatic respiratory viruses, suggesting that nasopharyngeal S pneumoniae proliferation is driven by cumulative mixed-viral co-infections. The effect of LAIV on pneumococcal density is probably similar to other respiratory viral infections in children. Our findings provide reassurance for the use of LAIV to expand influenza vaccine programmes in low-income to middle-income country populations with high pneumococcal carriage. FUNDING: Wellcome Trust.