New England Journal of Medicine DECEMBER 11, 2025

High-Dose Influenza Vaccine Effectiveness against Hospitalization in Older Adults (DANFLU-2)

The DANFLU-2 Investigators

Bottom Line

In this large-scale, registry-based, pragmatic randomized trial among older adults, high-dose inactivated influenza vaccine did not significantly reduce the primary composite endpoint of hospitalization for influenza or pneumonia compared to standard-dose vaccine.

Key Findings

1. The primary endpoint of hospitalization for influenza or pneumonia occurred in a similar proportion of participants in both groups, failing to demonstrate superiority for the high-dose vaccine (relative vaccine effectiveness 43.6% for influenza-specific hospitalizations but 0.5% for pneumonia hospitalizations).
2. Hospitalization for any cause was similar between groups, occurring in 9.38% of the high-dose group and 9.58% of the standard-dose group (relative effectiveness 2.1%; 95.2% CI, -0.1 to 4.3).
3. Mortality from any cause was nearly identical between the high-dose (0.67%) and standard-dose (0.66%) groups (relative effectiveness -2.5%; 95.2% CI, -11.6 to 5.9).
4. The incidence of serious adverse events was comparable between the high-dose and standard-dose vaccine groups.

Study Design

Design
RCT
Open-Label
Sample
332,438
Patients
Duration
Through May 31 of each influenza season
Median
Setting
Multicenter, Denmark
Population Older adults aged 65 years or older
Intervention High-dose inactivated influenza vaccine
Comparator Standard-dose inactivated influenza vaccine
Outcome Hospitalization for influenza or pneumonia occurring from 14 days after vaccination through May 31 of the following year

Study Limitations

The study was open-label, which may introduce ascertainment bias in the reporting of outcomes despite the use of administrative registries.
The primary composite outcome was largely driven by pneumonia hospitalizations, for which the high-dose vaccine showed no benefit, potentially obscuring benefits for influenza-specific outcomes.
The findings are based on observational registry data linked to a randomized trial, which, while pragmatic, may be subject to residual confounding if not fully captured by the registries.
Results may vary by influenza season intensity and circulating strain dominance, potentially limiting generalizability across all future seasons.

Clinical Significance

While high-dose influenza vaccines are often preferred based on immunogenicity and observational data, this large-scale pragmatic trial provides critical evidence that the high-dose vaccine did not reduce the overall burden of hospitalizations for influenza or pneumonia in the older population, suggesting that the clinical impact may be more nuanced than previously inferred from surrogate markers.

Historical Context

The efficacy of standard-dose influenza vaccines is known to decline with age due to immunosenescence, leading to the development of enhanced vaccines including high-dose (four-fold antigen) and adjuvanted formulations. Historically, regulatory approval and clinical recommendations for high-dose vaccines have relied on superiority in immunogenicity and observational effectiveness studies rather than definitive large-scale randomized trials targeting severe clinical outcomes, making the DANFLU-2 trial a landmark assessment of these public health strategies.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

What is the physiological basis for developing high-dose influenza vaccines specifically for the geriatric population compared to standard-dose formulations used in younger adults?

Key Response

Older adults experience immunosenescence, characterized by a decline in both innate and adaptive immune responses. This leads to lower antibody titers and reduced T-cell activation following standard vaccination. High-dose vaccines contain four times the antigen (60 μg vs 15 μg of hemagglutinin per strain) to provoke a more robust immune response and overcome this age-related diminished vaccine efficacy.

Resident
Resident

Despite the non-significant primary outcome in the DANFLU-2 trial, how should clinicians reconcile these findings with current CDC/ACIP recommendations regarding vaccine selection for patients over 65?

Key Response

Current ACIP guidelines preferentially recommend the use of high-dose, adjuvanted, or recombinant influenza vaccines for adults ≥65 years. While DANFLU-2 did not show a statistically significant reduction in its broad composite primary endpoint (hospitalization for influenza or pneumonia), earlier landmark trials like FIM12 showed a 24% higher efficacy for high-dose over standard-dose against laboratory-confirmed influenza. Residents should recognize that pragmatic trials with broad endpoints often suffer from 'outcome dilution' compared to efficacy trials with specific virologic endpoints.

Fellow
Fellow

The DANFLU-2 trial used a 'pragmatic, registry-based' design. How does the choice of 'hospitalization for pneumonia or influenza' as a primary composite endpoint, rather than laboratory-confirmed influenza infection, impact the study's power and the likelihood of observing a significant difference?

Key Response

Using broad ICD-10 codes for pneumonia and influenza increases the 'noise-to-signal' ratio. Because many pneumonia hospitalizations are caused by pathogens other than influenza (RSV, pneumococcus, etc.), a vaccine that only prevents influenza will have its measurable impact diluted. This 'outcome dilution' requires a much larger sample size to detect a difference and may mask the true efficacy of the vaccine against its specific target (the influenza virus).

Attending
Attending

Given the results of DANFLU-2, which demonstrated no significant difference in a Danish population with high baseline healthcare access, how should we weigh 'pragmatic' trial results against 'explanatory' trial results when making institutional formulary decisions for high-dose vaccines?

Key Response

Pragmatic trials like DANFLU-2 reflect real-world effectiveness in specific healthcare systems, while explanatory trials (like the original NEJM DiazGranados study) assess biological efficacy under controlled conditions. The lack of significance in DANFLU-2 may reflect a low-incidence influenza season or the high quality of supportive care in Denmark, suggesting that the 'value-add' of high-dose vaccines may vary significantly based on local epidemiology and healthcare infrastructure.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

DANFLU-2 utilized a decentralized, registry-based randomization approach. Discuss the potential for selection bias or 'healthy-vaccinee' bias in this pragmatic design compared to a traditional double-blinded phase III RCT.

Key Response

In registry-based pragmatic trials, the lack of stringent inclusion/exclusion criteria improves generalizability but introduces challenges. If the 'pragmatic' nature allowed for non-random patterns in who sought vaccination or if the registry failed to capture specific baseline comorbidities accurately, residual confounding could occur. Furthermore, the lack of blinding (though often mitigated by objective registry outcomes) can influence healthcare-seeking behavior or physician admitting thresholds if they are aware of the vaccine dose administered.

Journal Editor
Journal Editor

If you were reviewing the DANFLU-2 manuscript, why would you be cautious about the authors' emphasis on secondary outcomes (such as specific reductions in laboratory-confirmed influenza) when the primary endpoint failed to reach statistical significance?

Key Response

From an editorial standpoint, secondary outcomes are considered hypothesis-generating when the primary endpoint is non-significant. Emphasizing nominal p-values in secondary analyses without rigorous adjustment for multiple comparisons (alpha-spending) risks Type I error (false positives). A tough reviewer would flag 'cherry-picking' if the discussion focused on a 60% reduction in lab-confirmed influenza while the primary composite endpoint was null.

Guideline Committee
Guideline Committee

Should the DANFLU-2 results prompt a revision of the 'preferential recommendation' for high-dose influenza vaccines in the elderly, and how does this study fit into the GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) framework for current guidelines?

Key Response

One study, especially one with a broad non-specific endpoint, rarely overrules a body of evidence. Under the GRADE framework, DANFLU-2 would be viewed as 'moderate-to-high' quality evidence, but its results must be meta-analyzed with existing data. Because previous meta-analyses (e.g., Lee et al., 2017) consistently show a benefit against lab-confirmed influenza and influenza-related complications, the preferential recommendation remains stable, though DANFLU-2 highlights the need for more data on whether this translates to a reduction in all-cause pneumonia hospitalizations across different healthcare systems.

Clinical Landscape

Noteworthy Related Trials

2014

FIM12 Trial

n = 31,989 · NEJM

Tested

High-dose trivalent inactivated influenza vaccine

Population

Adults 65 years of age or older

Comparator

Standard-dose trivalent inactivated influenza vaccine

Endpoint

Laboratory-confirmed influenza illness

Key result: High-dose vaccine was 24.2% more effective than standard-dose vaccine in preventing laboratory-confirmed influenza in older adults.
2018

IVV-65+ Trial

n = 14,560 · JAMA

Tested

High-dose versus standard-dose inactivated influenza vaccine

Population

Community-dwelling elderly adults

Comparator

Standard-dose inactivated influenza vaccine

Endpoint

Influenza-related respiratory hospitalizations

Key result: The high-dose vaccine significantly reduced influenza-related hospitalizations compared to the standard-dose vaccine.
2021

NIV (NIV- Efficacy Study

n = 5,683 · Lancet Infect Dis

Tested

Adjuvanted quadrivalent influenza vaccine

Population

Adults 65 years of age or older

Comparator

Non-adjuvanted quadrivalent influenza vaccine

Endpoint

Relative vaccine efficacy against laboratory-confirmed influenza

Key result: Adjuvanted vaccine demonstrated higher efficacy in preventing influenza-like illness compared to the non-adjuvanted version in older adults.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis