Hearing loss in Australian First Nations children at 6-monthly assessments from age 12 to 36 months: Secondary outcomes from randomised controlled trials of novel pneumococcal conjugate vaccine schedules

Background In Australian remote communities, First Nations children with otitis media (OM)-related hearing loss are disproportionately at risk of developmental delay and poor school performance, compared to those with normal hearing. Our objective was to compare OM-related hearing loss in children randomised to one of 2 pneumococcal conjugate vaccine (PCV) formulations. Methods and findings In 2 sequential parallel, open-label, randomised controlled trials (the PREVIX trials), eligible infants were first allocated 1:1:1 at age 28 to 38 days to standard or mixed PCV schedules, then at age 12 months to PCV13 (13-valent pneumococcal conjugate vaccine, +P) or PHiD-CV10 (10-valent pneumococcal Haemophilus influenzae protein D conjugate vaccine, +S) (1:1). Here, we report prevalence and level of hearing loss outcomes in the +P and +S groups at 6-monthly scheduled assessments from age 12 to 36 months. From March 2013 to September 2018, 261 infants were enrolled and 461 hearing assessments were performed. Prevalence of hearing loss was 78% (25/32) in the +P group and 71% (20/28) in the +S group at baseline, declining to 52% (28/54) in the +P groups and 56% (33/59) in the +S group at age 36 months. At primary endpoint age 18 months, prevalence of moderate (disabling) hearing loss was 21% (9/42) in the +P group and 41% (20/49) in the +S group (difference -19%; (95% confidence interval (CI) [-38, -1], p = 0.07) and prevalence of no hearing loss was 36% (15/42) in the +P group and 16% (8/49) in the +S group (difference 19%; (95% CI [2, 37], p = 0.05). At subsequent time points, prevalence of moderate hearing loss remained lower in the +P group: differences -3%; (95% CI [-23, 18], p = 1.00 at age 24 months), -12%; (95% CI [-30, 6], p = 0.29 at age 30 months), and -9%; (95% CI [-23, 5], p = 0.25 at age 36 months). A major limitation was the small sample size, hence low power to reach statistical significance, thereby reducing confidence in the effect size. Conclusions In this study, we observed a high prevalence and persistence of moderate (disabling) hearing loss throughout early childhood. We found a lower prevalence of moderate hearing loss and correspondingly higher prevalence of no hearing loss in the +P group, which may have substantial benefits for high-risk children, their families, and society, but warrant further investigation.Trial registration ClinicalTrials.gov NCT01735084 and NCT01174849 Amanda Leach and colleagues compare two pneumococcal conjugate vaccines to examine the impact on hearing loss Author summary Why was this study done? Chronic otitis media (OM) (middle ear infection) causes hearing loss, which is linked to developmental delay, poor school readiness, attendance, and performance. Almost every Australian First Nations child living in remote regions experiences chronic OM throughout their formative early years. Vaccines effective against OM pathogens, namely, pneumococcus and nontypeable , could reduce chronic early childhood hearing loss and change the trajectory of disadvantage. Two pneumococcal conjugate vaccines with different formulations potentially protect from infection by either 13 pneumococcal strains (PCV13), or 10 of these 13 pneumococcal strains plus all nontypeable . (PHiD-CV10). Our hypothesis was that PHiD-CV10 would be superior to PCV13 in preventing hearing loss. No study had compared PCV13 and PHiD-CV10 in head-to-head or combination schedules. What did the researchers do and find? We designed 2 consecutive (primary course then booster dose) randomised controlled trials of head-to-head and combination PCV13 and PHiD-CV10 schedules. Audiologists assessed hearing at 6-monthly visits from age 12 to 36 months. Moderate (disabling) hearing loss was detected in 42% (25/60), 32% (29/91), 34% (29/86), 33% (37/111), and 19% (22/113) children at ages 12, 18, 24, 30, and 36 months, respectively. Hearing Australia's recommendation for hearing assistance was met in around 70% children at ages 12 (67%, 40/60), 18 (69%, 63/91), and 24 months (71%, 61/86), respectively. Contrary to our hypothesis, PHiD-CV10 vaccine was not superior to PCV13 in reducing hearing loss. At primary endpoint age 18 months, the prevalence of moderate (disabling) hearing loss was halved in the PCV13 group (21% (9/42) compared to 41% (20/49) in the PHiD-CV10 group, and the prevalence of no hearing loss was double (36% and 16%, respectively). At ages 24, 30, and 36 months, this trend continued but differences were smaller. What do these findings mean? The 19% lower prevalence of moderate hearing loss and 19% higher prevalence of no hearing loss in the PCV13 group at age 18 months is potentially highly clinically significant, particularly as the trend appears to be sustained, albeit smaller. Our study emphasises the importance of innovative and targeted research designs that meet the needs of high-risk populations and are independent of the pharmaceutical industry. Pneumococcal conjugate vaccine trials should evaluate impacts on child development. The small sample size creates uncertainty about the plausible difference in hearing between the two vaccines. We are 95% confident that the true difference is between 1% and 37%.