A Bayesian re-analysis of the DANFLU-1 trial

Area of Science:

• Gerontology
• Vaccinology
• Biostatistics

Background:

• The DANFLU-1 trial evaluated high-dose inactivated influenza vaccine (HD-IIV) versus standard-dose (SD-IIV) in adults aged 65-79.
• Previous findings indicated HD-IIV was associated with reduced mortality and hospitalization for influenza or pneumonia compared to SDIV.

Purpose of the Study:

• To conduct a post-hoc Bayesian re-analysis of the DANFLU-1 trial data.
• To probabilistically assess the relative vaccine effectiveness (rVE) of HD-IIV versus SD-IIV using various prior assumptions.

Main Methods:

• Bayesian log-binomial regression models were applied to data from 12,477 adults (65-79 years) randomized to HD-IIV or SDIV.
• The analysis incorporated neutral non-informative, evidence-based, and neutral skeptical priors.
• Relative vaccine effectiveness (rVE) and posterior probabilities for benefit thresholds (>0%, >10%, >20%) were estimated.

Main Results:

• Consistent results were observed across all priors.
• There was a >95% probability of any benefit (rVE > 0%) for all-cause mortality and pneumonia/influenza hospitalization.
• Probabilities of rVE > 10% for pneumonia/influenza hospitalization and all-cause mortality were high (≥91.1% and ≥93.2% across priors).

Conclusions:

• The Bayesian re-analysis robustly supports the clinical benefit of HD-IIV in older adults.
• High probabilities of effectiveness were demonstrated for all-cause mortality and pneumonia/influenza hospitalization.
• Further investigation in adequately powered trials is warranted.