Influenza Antibody Levels Associated with Laboratory-Confirmed Influenza in a Test-Negative Study Design, US Flu VE Network, November 2018-May 2019

In a 2018–2019 US test-negative study, higher serum antibody titers against circulating influenza viruses, particularly neuraminidase inhibition (NAI) and microneutralization (MN) titers, were significantly associated with reduced odds of laboratory-confirmed influenza infection in adults.

The Gist

Imagine your immune system as a highly trained security team guarding a fortress (your body) against a sneaky intruder: the flu virus. This paper is like a detective report from the 2018–2019 flu season, investigating how well this security team was doing its job when people actually got sick.

Here is the story of what the researchers found, broken down into simple concepts.

The Setup: The "Test-Negative" Detective Game

Usually, to see if a vaccine works, scientists give it to some people and a fake shot to others, then wait to see who gets sick. But that takes a long time and is expensive.

Instead, these researchers used a clever trick called a "Test-Negative Study." Imagine a busy hospital emergency room. People come in with coughs, fevers, and sore throats.

• The Suspects: Some of these people actually have the flu (the "Cases").
• The Innocent: Others have the flu symptoms but don't have the flu virus (the "Controls"). They might have a cold or allergies.

The researchers asked a simple question: "If you look at the security team's 'wanted posters' (antibodies) in the blood of the people who got sick versus those who didn't, is there a difference?"

The Security Team's Weapons: Three Types of Shields

The researchers looked at three different types of "weapons" the body uses to fight the flu, using three different lab tests:

1. The Net (Hemagglutination Inhibition - HI): This is like a net that tries to catch the virus before it can enter your cells. It's the most common shield people talk about.
2. The Sniper (Microneutralization - MN): This is a more precise weapon that neutralizes the virus directly. It's like a sniper taking out the virus before it can do damage.
3. The Jammer (Neuraminidase Inhibition - NAI): This doesn't stop the virus from entering, but it stops the virus from escaping your cells to infect others. It's like jamming the escape hatch so the virus gets stuck inside.

The Big Discovery: "The Right Shield Matters"

The researchers found something very important: It's not just about having antibodies; it's about having the right antibodies for the specific virus you are fighting.

• The "Old Map" Problem: The flu vaccine is made based on a "map" of what the virus might look like. Sometimes, the actual virus circulating in the community is a slightly different version (a "clade" or a new disguise).
• The Finding: When the researchers tested people's blood, they found that having high levels of antibodies against the exact virus circulating in the community (the "real" virus) was a much better predictor of staying healthy than having antibodies against the "old map" (the vaccine reference virus).

Think of it like this: If the burglar is wearing a red mask, having a security guard trained to spot a blue mask won't help much. You need a guard trained to spot the red mask.

The Results: How Much Protection Did They Have?

The study looked at two main types of flu: H1N1 and H3N2.

• For H1N1: People with higher levels of the "Net" (HI) and "Jammer" (NAI) antibodies were significantly less likely to get sick. Every time their antibody levels doubled, their odds of getting sick dropped by nearly half.
• For H3N2: This was the tricky one. The virus had changed its disguise (drifted). Even though vaccinated people had some antibodies, they weren't high enough or specific enough to stop this new version. This explains why the vaccine wasn't very effective against H3N2 that year. However, people who did have high levels of the "Sniper" (MN) and "Jammer" (NAI) antibodies against the actual circulating virus were much safer.

The Takeaway: Why This Matters

This study tells us two big things:

1. The "Gold Standard" isn't always enough: Just because a vaccine triggers an immune response doesn't mean it will stop the current flu strain if that strain has changed its disguise. We need to measure antibodies against the real virus, not just the one we guessed at months ago.
2. The "Jammer" is a Hero: The study highlighted that the "Jammer" antibodies (NAI) are a crucial part of the defense. Even if the "Net" (HI) fails to catch the virus, the "Jammer" can still stop it from spreading, offering a second line of defense.

In a nutshell: This research is like upgrading a security system. It tells us that to protect a city from a specific type of thief, we shouldn't just rely on a generic "thief" description. We need to know exactly what the thief looks like right now and train our guards to spot that specific disguise. If we do that, we can better predict who will stay safe and who might get sick, helping us make better vaccines for the future.

Technical Summary

1. Problem Statement

Seasonal influenza vaccine effectiveness (VE) varies significantly by season, circulating viral strains, and host immunity. A major challenge in vaccine development and evaluation is establishing reliable serologic correlates of protection (antibody thresholds that predict immunity), particularly when circulating viruses are antigenically drifted from vaccine reference strains.

While randomized trials and cohort studies have traditionally been used to determine these correlates, they often rely on post-vaccination antibody titers measured against vaccine reference strains, which may not accurately reflect protection against circulating, drifted viruses. There is a need for observational methods that can assess antibody levels close to the time of exposure in real-world settings to better understand the relationship between specific antibody titers (against Hemagglutinin and Neuraminidase) and protection against symptomatic, medically-attended influenza.

2. Methodology

The study utilized a Test-Negative Design (TND) within the US Flu Vaccine Effectiveness (VE) Network during the 2018–2019 influenza season (November 2018 – May 2019).

• Study Population: 6,323 ambulatory outpatients (aged ≥18 years) presenting with acute respiratory illness within 7 days of symptom onset across five US states.
• Serologic Substudy: A subset of 305 patients provided acute-phase serum specimens.
  • Cases: 175 patients with laboratory-confirmed influenza (112 Influenza A(H1N1)pdm09; 63 Influenza A(H3N2)).
  • Controls: 130 patients who tested negative for influenza (test-negative controls).
• Virological Characterization: Respiratory specimens were tested via RT-PCR. Viral clades were identified via whole-genome sequencing at the CDC.
  • A(H1N1)pdm09 cases were predominantly clade 6B.1A.
  • A(H3N2) cases were predominantly clade 3C.3a.
• Serological Assays: Three distinct assays were performed on acute-phase sera to measure antibody titers against both vaccine reference strains and circulating clade-specific strains:
  1. Hemagglutination Inhibition (HI): Measured antibodies against Hemagglutinin (HA).
  2. Microneutralization (MN): Used for A(H3N2) due to reduced hemagglutination capability of certain clades; measured neutralizing antibodies.
  3. Neuraminidase Inhibition (NAI): Measured antibodies against Neuraminidase (NA) using reassortant viruses (rH6N1/rH6N2) to avoid HA interference.
• Statistical Analysis:
  • Geometric Mean Titers (GMT) and proportions of patients with titers ≥10 and ≥40 were compared between cases and controls.
  • Logistic Regression: Used to estimate the odds of influenza infection associated with increasing antibody titers. Models were adjusted for vaccination status and time since the start of the flu season.
  • Seroconversion: Calculated for a subset of patients with paired acute and convalescent samples (21+ days post-onset) to measure the immune response to infection.

3. Key Contributions

• Application of TND for Correlates of Protection: This study demonstrates the utility of the test-negative design not just for VE estimation, but for identifying antibody thresholds associated with protection in a real-world setting. It treats acute-phase titers as a proxy for pre-exposure immunity.
• Antigenic Specificity: The study highlights the critical importance of using circulating viral clades rather than vaccine reference strains in serological assays. It found that antibodies measured against the actual infecting virus were more strongly associated with protection than those measured against the vaccine strain.
• Role of Neuraminidase (NA): The study provides robust evidence that NAI antibodies are independently associated with protection against both A(H1N1)pdm09 and A(H3N2), even after controlling for HI/MN titers. This supports the inclusion of NA in vaccine evaluation and the search for broader correlates of protection.

4. Key Results

General Findings

• Higher acute-phase antibody titers (HI, MN, and NAI) against circulating viruses were significantly associated with a lower likelihood of testing positive for influenza.
• A(H1N1)pdm09:
  • Cases had significantly lower GMTs for HI and NAI compared to controls.
  • Odds Reduction: Each two-fold increase in HI titer against cell-propagated A(H1N1) reduced the odds of infection by 44% (OR 0.56). Each two-fold increase in NAI titer reduced odds by 54% (OR 0.46).
  • Seroconversion rates were high (70% for HI, 80% for NAI) among infected cases.
• A(H3N2) (Clade 3C.3a):
  • Cases had significantly lower MN titers against the circulating 3C.3a virus compared to controls. No significant difference was found against the egg-propagated vaccine strain (3C.2a.1).
  • Odds Reduction: Each two-fold increase in MN titer against circulating A(H3N2) reduced the odds of infection by 49% (OR 0.51). Each two-fold increase in NAI titer reduced odds by 28% (OR 0.72).
  • Despite vaccination, overall MN titers against A(H3N2) were low, correlating with the observed lack of vaccine effectiveness against A(H3N2) in this season.

Vaccination Impact

• Among uninfected controls, vaccinated individuals generally had higher antibody titers than unvaccinated individuals.
• However, for A(H3N2), the absolute titers remained low even in vaccinated individuals, suggesting that the vaccine-induced immunity was insufficient to prevent infection by the drifted 3C.3a clade.

5. Significance and Implications

• Refining Correlates of Protection: The study confirms that serologic correlates of protection must be strain-specific. Using vaccine reference strains may underestimate or misrepresent the actual protection against circulating, drifted viruses.
• Importance of Neuraminidase: The independent protective association of NAI titers suggests that future vaccine evaluations should consider NA immunity, potentially leading to vaccines that offer broader protection across clades.
• Methodological Validation: The findings validate the use of test-negative studies to estimate correlates of protection, offering a more efficient and ecologically valid alternative to large-scale randomized trials, especially for antigenically drifted seasons.
• Vaccine Strain Selection: The data supports the need for regular updates to serology antigens to match circulating clades, which is crucial for accurate vaccine strain selection and immunogenicity monitoring.

Limitations Noted: The serologic subset was not randomly selected, pediatric and hospitalized populations were excluded, and acute-phase titers in some cases might reflect early anamnestic responses (though literature suggests this is minimal in the first week).

Conclusion: Higher influenza antibody titers against circulating viruses are strongly associated with protection against medically attended influenza. This approach provides a practical framework for evaluating vaccine performance and identifying immune thresholds in the context of antigenic drift.