2024/25 end-of-season KP.2 vaccine effectiveness against COVID-19 hospitalization in older adults: a test-negative study in Quebec, Canada

A test-negative study in Quebec involving nearly 50,000 older adults found that the 2024/25 KP.2 vaccine provided a meaningful but short-lived 34% overall effectiveness against COVID-19 hospitalization, with protection declining from 43% within eight weeks to negligible levels by 32 weeks post-vaccination.

The Gist

Imagine your immune system is a castle, and the COVID-19 virus is an army of invaders trying to break through the gates. Every year, the virus changes its armor and weapons (new variants), so the castle needs to update its defenses.

This study is like a report card on how well the 2024/25 "KP.2" vaccine worked as a shield for older adults in Quebec, Canada, against getting sick enough to end up in the hospital.

Here is the breakdown of what happened, using some everyday analogies:

1. The Setup: A Seasonal Update

Just like you buy a new winter coat every year because the old one is worn out or the weather changed, scientists updated the COVID-19 vaccine to match the newest version of the virus (the KP.2 variant).

• The Goal: To stop older adults (60+) from getting so sick they need to go to the hospital.
• The Timing: They rolled out the vaccine in two waves: a big Fall campaign (like stocking up for winter) and a smaller Spring campaign (a top-up for the very elderly).

2. The Experiment: The "Test-Negative" Detective Game

How do researchers know if the vaccine worked? They played a game of "detective" using hospital records.

• The Scene: They looked at thousands of older adults who went to the hospital because they felt sick (coughing, fever, etc.).
• The Clue: They checked who had the virus (the "Cases") and who didn't have the virus but was still sick with something else (the "Controls").
• The Question: "Did the people who ended up in the hospital with COVID have the vaccine, or did the people who were not in the hospital with COVID have the vaccine?"
• The Result: If the sick people were mostly unvaccinated, the vaccine is working. If the sick people were mostly vaccinated, the vaccine isn't working.

3. The Findings: A Strong Shield That Fades

The study found that the vaccine was a very effective shield, but it was a "short-term" shield.

• The Fresh Shield (First 2 Months): Right after getting the shot, the vaccine was about 43% effective. Think of this like a brand-new, high-tech raincoat. It keeps you mostly dry. It reduced the risk of ending up in the hospital by about one-third to one-half.
• The Waning Shield (Months 3–6): As time passed, the "raincoat" started to get a few holes in it. By 3 to 6 months after the shot, the protection dropped to about 30%. It was still helping, but not as much.
• The Tattered Shield (7+ Months): By the time 7 months (32 weeks) had passed, the protection was basically gone. The shield had worn out completely. If you got the vaccine in the fall, by the following spring, it offered almost no protection against hospitalization.

4. The Spring Campaign: A Small Top-Up

The researchers also looked at the Spring campaign.

• The Problem: Very few people got the spring shot (only about 2% of the control group). It was like trying to fix a leaky roof with a tiny patch when the whole roof needs replacing.
• The Result: Even with the low numbers, the spring shot seemed to work just as well as the fresh fall shot at the time it was given. However, because so few people took it, it didn't make a huge dent in the overall numbers.

5. The Big Picture: Why This Matters

The study teaches us three main lessons:

1. It Works, But Briefly: The vaccine is great at stopping older adults from getting hospitalized, but that protection doesn't last a whole year. It's more like a "seasonal flu shot" that needs to be refreshed often.
2. The Virus Keeps Changing: The virus kept swapping its "armor" (new variants like XEC, KP.3, and LP.8.1) throughout the year. The vaccine was designed for one version (KP.2), but the virus kept evolving, which might have made the shield less effective over time.
3. We Need a New Strategy: Since the protection fades after about 6 months, and the virus can strike in the summer or winter, we might need to think about giving vaccines more often (like every 6 months) to high-risk groups, rather than just once a year.

In a nutshell: The 2024/25 vaccine was a good "emergency brake" that stopped many older adults from crashing into the hospital, but the brake pads wore out quickly. To stay safe, we likely need to hit the brakes again sooner than we thought.

Technical Summary

1. Problem Statement

The study addresses the critical need to evaluate the real-world effectiveness (VE) and durability of protection provided by the updated 2024/25 KP.2-adapted mRNA COVID-19 vaccine. While early-season data suggested moderate protection against hospitalization, evidence regarding protection beyond six months was limited. Furthermore, no data existed on the effectiveness of the spring 2025 vaccination campaign in Canada. The study aims to quantify the VE against COVID-19 hospitalization in adults aged ≥60 years in Quebec, Canada, over a full ten-month season (fall and spring), and to assess how protection wanes over time and across different circulating subvariants (KP.3, XEC, LF.7, LP.8.1, etc.).

2. Methodology

• Study Design: A population-based test-negative case-control study.
• Setting & Population: Residents of Quebec, Canada, aged ≥60 years who were tested for COVID-19-compatible symptoms in acute-care hospitals.
• Timeframe: October 13, 2024 (Epi-week 2024-42) to August 23, 2025 (Epi-week 2025-34).
• Data Sources: Linked administrative databases including provincial vaccination records, hospital admissions, chronic disease registries, and laboratory testing results.
• Case Definition: SARS-CoV-2 positive tests followed by a hospital admission (≥24 hours) with a COVID-19 diagnosis.
• Control Definition: SARS-CoV-2 negative tests (with or without hospitalization) from individuals with similar symptoms.
• Exclusion Criteria:
  • Immunocompromised individuals.
  • Residents of long-term care facilities (due to differing hospitalization protocols).
  • Tests within 90 days of a prior positive result (to exclude reinfections).
  • Vaccination <7 days prior to specimen collection.
  • Invalid vaccination dates or intervals.
• Statistical Analysis: Logistic regression models were used to calculate adjusted odds ratios (OR) comparing vaccination status between cases and controls. VE was calculated as $(1 - \text{adjusted OR}) \times 100$.
• Covariates: Models adjusted for sex, age, comorbidities, place of residence (home vs. private senior housing), and epidemiological week.
• Stratification: Analyses were conducted for the entire season, separately for fall and spring campaigns, by time since vaccination (in 8-week increments), by number of doses, and by calendar period.

3. Key Contributions

• Comprehensive Seasonal Evaluation: This is one of the first studies to provide end-of-season VE estimates spanning a full ten-month period (fall through late summer), covering both the primary fall campaign and the secondary spring campaign.
• Waning Analysis: It provides granular data on the duration of protection, specifically tracking effectiveness from <8 weeks post-vaccination up to 32+ weeks.
• Spring Campaign Data: It offers the first reported VE estimates for the 2025 spring vaccination campaign in Canada, a period often characterized by lower viral circulation and uptake.
• Subvariant Context: While individual genotyping was not available, the study contextualizes VE results against the known circulation of multiple subvariants (KP.3, XEC, LF.7, LP.8.1, NB.1.8.1, XFG) during the study period.

4. Key Results

• Study Population: Out of 157,670 tests, 49,949 participants (43%) were eligible for the analysis. Of these, 43% were vaccinated with the KP.2 vaccine.
  • Vaccine Types: 38.2% received Pfizer-BioNTech (BNT162b2) and 61.8% received Moderna (mRNA-1273).
  • Uptake: Fall vaccination coverage was higher (41.1% of controls) compared to the spring campaign (2.3% of controls).
• Overall Effectiveness:
  • The overall VE against hospitalization for the 2024/25 season was 34% (95% CI: 28–40).
  • Median time since vaccination was 16 weeks (IQR: 9–24).
• Temporal Waning:
  • <8 weeks post-vaccination: VE was 43%.
  • 8–31 weeks: VE declined to approximately 30%.
  • ≥32 weeks: Protection became negligible (statistically indistinguishable from zero).
• Campaign Specifics:
  • Fall Campaign: VE was 38% at a median of 12 weeks post-vaccination.
  • Spring Campaign: VE was 38% (95% CI: 3–60) at a median of 7 weeks. Despite low sample size and uptake, the point estimate was similar to the early fall estimates.
• Age Stratification: VE estimates were lower and less precise in the 60–69 age group compared to older groups (70–79 and ≥80), potentially due to lower coverage or residual confounding.

5. Significance and Implications

• Short-Lived Protection: The study confirms that while KP.2-adapted vaccines provide meaningful protection against severe outcomes (hospitalization) in older adults, this protection is short-lived, waning significantly after 3–4 months and becoming negligible by 6–7 months.
• Policy Implications: The findings support the necessity of repeated vaccination strategies, potentially including spring doses for high-risk groups, to maintain protection against unpredictable SARS-CoV-2 seasonality and variant emergence.
• Variant Evolution: The decline in VE may be attributed to both immunological waning and the emergence of immune-evasive subvariants (such as LP.8.1 late in the season), although the study could not disentangle these factors due to a lack of individual genotyping.
• Global Context: The results align with US data (showing ~45% early VE) but are lower than some European estimates using JN.1-adapted vaccines, suggesting potential differences in vaccine formulation efficacy against circulating lineages in North America versus Europe.

Conclusion: The 2024/25 KP.2 vaccine effectively reduced COVID-19 hospitalization risk by approximately one-third in the short term for older adults in Quebec, but protection waned rapidly, becoming negligible by the seventh month. This underscores the need for ongoing surveillance and potentially more frequent booster strategies for vulnerable populations.