Vaccine Effectiveness of BNT162b2 Against Delta and Omicron Variants in Adolescents

We conducted a test-negative design study among individuals aged 12 to 17 years and SARS-CoV-2 tested (by real-time reverse transcription polymerase chain reaction) between November 22, 2021 (date of first Omicron detection) and March 6, 2022 using methods detailed elsewhere.^1,2  We restricted our Delta outcomes to January 2, 2022 and earlier because of small numbers and limited circulation after this date. We linked and analyzed laboratory, vaccination, reportable disease, and health administrative data for the entire eligible population at ICES (formerly the Institute for Clinical Evaluative Sciences). We estimated VE against symptomatic infection and severe outcomes (ie, hospitalization or death) over time since second or third dose receipt (using discrete time intervals after a second dose and ≥7 days after a third dose, due to limited follow-up) for both Omicron and Delta, which were defined by using a combination of whole genome sequencing and S-gene target failure (SGTF) results, and dates (Supplemental Table 1).²  In a sensitivity analysis, we estimated VE against symptomatic Omicron infection before and after restrictions to polymerase chain reaction test eligibility were announced at the end of December 2021 because of overwhelming case volumes. We used multivariable logistic regression to compare the odds of vaccination in cases with symptomatic test-negative controls and calculated VE as a 1-adjusted odds ratio. Projects that use data collected by ICES under section 45 of Ontario’s Personal Health Information Protection Act, and use no other data, are exempt from research ethics board review.

We included 9902 Omicron cases (compared with 19 953 test-negative controls) and 502 Delta cases (compared with 17 930 test-negative controls) (Supplemental Table 2). For the Omicron analyses, 91% of tested subjects had received 2 doses and 1.3% had received 3 doses, with most (84.5%) having an interval between their first 2 doses of <8 weeks (Supplemental Table 3). VE was lower against symptomatic Omicron infection than against Delta and decreased more rapidly over time, from 51% (95% confidence interval [CI], 38% to 61%) 7 to 59 days after a second dose to 29% (95% CI, 17% to 38%) after 180 days, compared with 97% (95% CI, 94% to 99%) and 90% (95% CI, 79% to 95%) for the same intervals against symptomatic Delta infection (Fig 1, Supplemental Table 4). Overall, 2 dose VE against severe outcomes due to Omicron was 85% (95% CI, 74% to 91%) ≥7 days after a second dose, and estimates were similar over time. Although we were unable to assess 3-dose VE against severe Omicron outcomes or any Delta outcomes because of small numbers (with zero severe Delta events among vaccinated individuals), VE against symptomatic Omicron infection was 62% (95% CI, 49% to 72%) ≥7 days after a third dose. Results against symptomatic Omicron infection were similar before and after changes to test eligibility criteria (Supplemental Table 5 and Supplemental Fig 2).

VE against symptomatic Omicron infection wanes over time in adolescents given 2 doses of BNT162b2 and was higher in the initial period after a third dose. Although lower than observed against Delta, protection against severe outcomes appears to be maintained over time. The declining VE against infection is similar to results from England in which VE against Omicron and Delta in adolescents aged 16 to 17 years decreased from 76% and 93% after 7 to 13 days to 23% and 84% ≥70 days post second dose, respectively.³  In 1 United States study, 2 dose VE against Omicron hospitalization for those aged 12 to 15 and 16 to 17 years was 92% and 94% after 14 to 149 days compared with 73% and 88% after ≥150 days, respectively.⁴  This difference in VE over time since second dose was not statistically significant, whereas declines in VE against emergency department (ED) visits were. VE against ED visits increased after a third dose in those aged 16 to 17 years.⁴  In another United States study, VE against Omicron hospitalization was estimated at 40% among those aged 12 to 18 years, with a median interval of 162 days since vaccination.⁵  Limitations of this study included an inability to classify some specimens using SGTF or sequencing, changes to test eligibility over the study period, limited follow-up time after third doses, and potential unmeasured confounding between vaccinated and unvaccinated individuals. These results can inform third dose recommendations in adolescents because 2-dose protection against symptomatic Omicron infection is relatively low and wanes over time, whereas protection of a second dose against severe outcomes is higher. A third dose substantially improves protection against symptomatic Omicron infection in adolescents, but 3-dose VE is only moderate at ∼60% in the early period after vaccination, and the duration of this protection is unknown. Understanding the impact of a third dose on improved protection against severe outcomes will be important.

We would like to acknowledge Public Health Ontario for access to vaccination data from COVaxON, case-level data from the Public Health Case and Contact Management Solution (CCM), and coronavirus disease 2019 laboratory data, as well as assistance with data interpretation. We also thank the staff of Ontario’s public health units who are responsible for coronavirus disease 2019 case and contact management and data collection within CCM. We thank IQVIA Solutions Canada Inc. for use of their Drug Information File. The authors are grateful to the Ontario residents without whom this research would be impossible.

CI

confidence interval

ED

emergency department

SARS-CoV-2

severe acute respiratory syndrome coronavirus 2

SGTF

S-gene target failure

VE

vaccine effectiveness