Influenza Hemagglutinin Antibody Levels in the Elderly: Impact of Sex, Age, and Influenza/COVID-19 Vaccination Status During the 2021/2022 Epidemic Season in Warsaw, Poland

Introduction

The outbreak of SARS-CoV-2 infections in 2019 in China has changed government health policies worldwide. A new and unexpected threat has diverted attention from other infectious diseases. Although more than 676 million total cases and 6.8 million total deaths due to COVID-19 were reported from the beginning of the pandemic until October 2023, it is still not the deadliest infectious disease ever [1]. The “big three” are tuberculosis, estimated at 1880 million total infections and around 10 million new cases per year, malaria, and HIV/AIDS [2]. However, influenza viruses, affecting humans and animals (depending on the type), remain a global threat. Type A influenza virus presents high variability and mutates constantly, causing periodic pandemics.

Influenza was first mentioned by Hippocrates in 412 BC, and since then, it has caused many epidemics and at least several pandemics with dramatic life loss. The “Spanish flu” was responsible for about 500 million acute viral infection cases and 50 to 100 million deaths [3]. According to the World Health Organization (WHO), even now, seasonal influenza affects annually around 1 billion people, with 3 to 5 million cases of severe illness and 290 000 to 650 000 respiratory deaths. The most endangered are children under 5 years of age, pregnant women, adults with specific chronic medical conditions, including obesity, and older people, especially those with multimorbidity [4].

The world population is rapidly aging. By 2030, according to the WHO, 1 in 6 people worldwide will be 60 years old or older [5]. Simultaneously, we face the increase in multimorbidity, defined as the coexistence of 2 or more chronic diseases. Both age and some health problems, such as diabetes, chronic obstructive pulmonary disease, and cardiovascular diseases, worsen the course of influenza infection and cause additional problems, such as bacterial superinfections, severe pulmonary, nephrological, metabolic, and cardiovascular complications (sometimes requiring long-term treatment) and last, but not least, a huge organizational and financial burden for the healthcare systems around the world.

Global efforts to stop the spread of influenza are needed. An effective campaign requires gathering as much information as possible. Therefore, the role of the Global Influenza Surveillance and Response System, which has 151 National Influenza Centers in 129 countries, is crucial. Virological and epidemiological monitoring of the influenza virus helps to acquire all possible data valuable, also in inventing new drugs and typing the most effective vaccine for each season (that lasts from October until the end of September the following year). Common vaccination is the most effective method of reducing the number of infected, severity of influenza infections, serious health complications, death rate, and high social cost [6]. According to the recommendation of the Advisory Committee on Immunization Practices (2021/2022), in every epidemic season, influenza vaccination is recommended for all persons aged ≥6 months who do not have contraindications, optimally before the onset of influenza activity in the community. The effectiveness of vaccination varies in the population due to several factors, such as age, previous medical conditions, individual immune response activity, and the similarity between the circulating viruses and types administered in the vaccine [7].

Age-related changes in immune mechanisms (both in cellular and humoral immunity) cause a weakening of the effectiveness of vaccination. In the overall population during the typical influenza season, vaccination reduces the risk of acute infection by between 40% and 60%, while in the population over 65 years of age, the effect varies depending on the virus subtype, from 10.3% for influenza A subtype A(H3N2), 28.5% for influenza B, to 61.2% for influenza A subtype A(H1N1) pdm09 [8,9]. Such heterogeneous results require more investigation to establish the best possible composition of the vaccine and the best vaccination tactics in the geriatric population.

Therefore, it is crucial to measure the protective factor, defined as the percentage of people with a level of anti-hemagglutinin antibodies titer of at least 1: 40, which reduces the risk of influenza by 50% [10]. To establish this factor, a typical laboratory method of the hemagglutination inhibition assay in the patient’s serum is used [11]. Antihemagglutination antibodies in the sample block the binding of the surface glycoprotein called “hemagglutinin” to the sialic acid sites on the red blood cells. The protection factor was established for healthy adults and can be reached from vaccination and previous infection.

In Poland, special free influenza vaccinations for elderly adults are available. Unfortunately, the number of elderly patients vaccinated is still very low.

The aim of our research was to estimate the titer of anti-hemagglutinin antibodies among elderly patients during the epidemic season of 2021/2022, especially considering sex and age, and their response to previous vaccination (both against influenza and COVID-19).

Material and Methods

ETHICS:

This study was approved by the NIZP PZH-PIB Bioethics Committee, Warsaw, Poland, April 8, 2022 (reference No. 3/2022). According to the national legislation, informed consent was obtained from all individual participants in the study before blood samples were collected. Participant’s confidentiality was ensured according to the General Data Protection Regulation (Regulation (EU) 2016/679), a European Union regulation on information privacy in the European Union and the European Economic Area. No significant risk for the participants was expected.

STUDY DESIGN AND PATIENTS STUDIED:

Patients under the care of geriatric and general practitioners in Doctor’s Surgery NZOZ Nucleus, Warsaw, Poland, were included successively during the 2021/2022 influenza epidemic season.

The overall study population consisted of 256 geriatric patients aged 65 to 99, comprising 193 (75.4%) women and 63 (24.6%) men, with an average age of 78.7 years (women: 79.2±0.7 years, men: 77.0±1.1 years). The exclusion criteria were lack of informed consent, severe active infectious disease, exacerbation of chronic diseases (eg, heart failure, respiratory failure), acute coronary syndrome, active stroke, and dementia.

The patients’ vaccination status was as follows: 229 seniors (89.4%), comprised of 174 (75.9%) women and 55 (24.1%) men, were vaccinated against COVID-19 by Pfizer or Moderna vaccines;

78 (30.5%) patients were vaccinated against influenza with a quadrivalent influenza vaccine prepared by Sanofi Pasteur for the epidemic season 2021/2022. Of this group, 54 (69.2%) patients were women and 24 (30.8%) were men.

A total of 77 patients were vaccinated both against COVID-19 and influenza. This group included 53 women (68.83%) and 24 (31.17%) men.

A total of 26 (10.2%) patients, including 18 (69.23%) women and 8 (30.77%) men, were not vaccinated against either COVID-19 or influenza.

In the examined population, 130 patients (50.8%) had cardiovascular disease, 83 (32.4%) had diabetes, 30 (11.7%) had cancer, 28 (20.9%) had respiratory system disease, and 10 (3.0%) had autoimmune disease.

METHODS:

The levels of antihemagglutinin antibodies against 4 antigens, A/Victoria/2570/2019(H1N1)pdm09, A/Cambodia/e0826360/2020(H3N2), B/Washington/02/2019 (B/Victoria lineage), B/Phuket/3073/2013 (B/Yamagata lineage) of the quadrivalent influenza vaccine (2021/2022) were determined in the serum of vaccinated and unvaccinated patients.

The patients’ venous blood samples were collected. Then, the anonymized sera samples were obtained in centrifugation, frozen at −30 °C, and sent to the Department of Influenza Research-National Influenza Center. Selected viruses obtained from the World Influenza Centre at the Francis Crick Institute, London, were propagated in chicken embryos in the National Influenza Centre, Warsaw. Then, the solution containing viruses was diluted to a titer of 1: 8 and incubated for 30 min at room temperature with 0.1% calcium salt, receptor destroying enzyme, turkey red blood cells, and patients’ serum samples in the hemagglutination inhibition (HAI) test. The titer of antihemagglutinin antibodies was determined and analyzed at levels of 0, 10, 20, 40, 80, 160, 320, and 640.

For statistical analysis, the following tests were used: Mann-Whitney, chi-square test, and Fisher test (for sample numbers that were too low, which cannot be estimated in the chi-square test). In all tests, the level of statistical significance was P≤0.05. For the patients with particular antibodies, the average level of antibodies (geometric mean titer [GMT]) and the percentage of people reaching the protective anti-hemagglutinin antibody titer (>40) were analyzed. Those parameters were determined along with the 95% confidence interval (95% CI) limits for geometric mean and fraction, respectively. The 95% CI for the geometric mean was calculated using the confidence interval for the arithmetic mean of the logarithms of the measured values.

Results

PREVALENCE OF ANTIBODIES:

The number and percentage of patients with antibodies differed depending on the subtype of influenza A and influenza B viruses of both the Victoria and Yamagata lines (Table 2). The lowest percentage of patients with antibodies, 59.4% (n=152), was recorded in the case of influenza A virus subtype A/Victoria/2570/2019(H1N1) pdm09. In the case of other viruses, these values were at a similar level: 96.1% (246 patients) for A/Cambodia/e0826360/2020(H3N2), 96.5% (247 patients) for B/Washington/02/2019 (B/Victoria lineage), and 98.4% (252 patients) for B/Phuket/3073/2013 (B/Yamagata lineage).

The highest GMT values were recorded for the A/(H3N2) subtype virus, at 110.9, and the lowest for the A/(H1N1) pdm09 subtype, at 35.4. The GMT for B/Victoria lineage was 44.3, and for the B/Yamagata lineage, 61.8.

That indicates the highest growth in antibodies titer was for A/Cambodia/e0826360/2020 (H3N2), while the lowest was for A/Victoria/2570/2019(H1N1)pdm09. Similarly, the protective level was the highest for influenza subtype A(H3N2) virus (224 patients [87.5%]), and the lowest was for influenza virus subtype A(H1N1)pdm09 (70 patients [27.3%]).

RELEVANCE OF VACCINATION:

We analyzed the GMT and protective levels in patients vaccinated against influenza and COVID-19. A total of 78 (30.5%) patients were vaccinated against influenza and 178 (69.5%) were unvaccinated. The data are shown in Table 3.

Taking into account people with non-zero antibody titers, the values were on average higher in the vaccinated group for all virus types, except for the A/Victoria/2570/2019(H1N1) pdm09, which was not statistically significant (GMT 40.0 vs 32.4). What seems to be important for this virus subtype is that among 78 patients vaccinated against influenza, 63 (80.8%) had the non-zero antibody titers and 15 (19.2%) showed no immune response to vaccination. For A/Cambodia/e0826360/2020(H3N2), only 3 patients (3.8%) showed no response, and for influenza type B, all the patients had antibodies.

Only 89 patients (50%) in the group of 178 unvaccinated patients had the non-zero titers for A/Victoria/2570/2019(H1N1)pdm09 virus subtype, while for the other subtypes, the number varied from 169 (94.9%) for B/Washington/02/2019(B/Victoria lineage) to 174 (97.8%) for B/Phuket/3073/2013(B/Yamagata lineage).

In the study group, the protective level was significantly higher in patients vaccinated against influenza (1-sided chi-square test), except for A/Cambodia/e0826360/2020(H3N2), at 89.7% vs 86.5%, as shown in Table 4. For the virus subtype A/H1N1/pmd09, the protective level of vaccinated patients was almost double that of unvaccinated patients (43.6% vs 20.2%).

A total of 77 patients (30.1% of 256 patients) were vaccinated against influenza and COVID-19, which influenced the results of GMT and protective levels for all 4 subtypes of influenza. As shown in Table 5, GMT values differed the most for A/H3N2 (116.8 for vaccinated vs 71.9 for unvaccinated; P=0.008). Among B/Victoria and B/Yamagata lineage, the differences were, respectively, 46.6 vs 27.9; P=0.007, and 65.9 vs 35.9; P=0.002. The result for A/H1N1 pmd09 was not statistically significant (35.8 vs 30.3).

Vaccination against COVID-19 also effects the protective levels, which are higher among the vaccinated, although the result for A/H1N1/pdm09 (28.8% vs 14.8%) was borderline significant (P=0.061) and A/H3N2 (87.8% vs 85.2%) was not statistically significant. The results for both B/lineages were much higher (64.2% vs 44.4%; P=0.023 and 78.6% vs 55.6%; P=0.004, respectively), as shown in Table 6.

RELEVANCE OF PERSONAL CHARACTERISTICS:

The study group included 193 women (75.4% of all patients) and 63 men (24.6%). Table 7 shows the GMT values, which vary depending on patient sex. For all subtypes, GMT values were higher among men than women, but the statistically significant difference could be observed only for B/Washington/02/2019/(B/Victoria lineage) (67.6 vs 38.4; P<0.001; a 2-sided Mann-Whitney test). Another comparison refers to the differences in protective levels among both sexes, as shown in Table 8.

As for GMT, protective levels were also higher among men for all subtypes of influenza viruses. However, the difference was statistically significant also only for B/Washington/02/2019 (74.6% vs 58.0%; P<0.001), while for A/Victoria/2570/2019(H1N1)pdm09 it was borderline significant (36.5%, vs 24.4%; P=0.060).

In this study, we also wanted to establish whether there is a difference between younger and older seniors. In the analyzed group, there were 138 (53.9%) people aged 65 to 79 years and 118 (46.1%) 80 years and older. GMT values for patients 65 to 79 years and older (aged 80+) were analyzed using the Mann-Whitney test, as shown in Table 9.

For all antibodies except A/Victoria/2570/2019(H1N1)pdm09 (30.5 vs 40.3), GMT values were higher among people aged ≥80 years, but there were no statistically significant differences. The values for A/H3N2 were 117.7 vs 105.5, for B/Victoria lineage, 74.4 vs 41.7, and for B/Yamagata lineage, 65.7 vs 58.6.

The results for protective level values were also not significant, which is visible in Table 10. However, it is surprising that the results for A/H1N1/pmd09 and A/H3N2 were higher for younger patients (29.7% vs 24.6% and 87.7% vs 87.3%, respectively).

Discussion

Vaccination against influenza is the most effective method of preventing illness cases, post-influenza multiorgan complications, and deaths, regardless of age. Other vital reasons for vaccination, especially among elderly adults, are the huge social impact of the disease, very high and measurable financial costs, and burden on the healthcare system [12,13]. However, the percentage of the vaccinated population in Poland remains low despite a large educational campaign and free-of-charge vaccines for people older than 75 years of age. The percentage of the vaccinated population in the country is growing very slowly, despite the convincing results of research conducted in cooperation with clinicians in 32 high-risk groups [14]. In general, based on the analysis conducted in epidemic seasons from 2016/2017 to 2021/2022, 15.83% of the population over 65 years of age is vaccinated against influenza (which is the highest result in Poland) [15].

In addition to the number of vaccinated elderly adults, another important issue is their response to the vaccination. In this study, we tried to answer the question of effectiveness of vaccination among geriatric patients and the factors that influenced it.

The analysis of the results indicate that in the sera of the tested patients, the lowest percentage of patients with antibodies was recorded for the influenza virus A/Victoria/2570/2019(H1N1)pdm09 subtype, which was 59.4%. It is worth noting that only this subtype showed a lack of immune response, in 15 (19.2%) patients after vaccination. The protective level was also the lowest and was reached only among 70 patients (27.3%). On the other hand, vaccination against influenza has increased mostly the number of patients with non-zero antibody titers for this subtype, from 50.0% among unvaccinated to 80.8% among vaccinated (Table 3). Subtype A/Cambodia/e0826360/2020(H3N2) showed the highest protective level (87.5%, 224 patients) and GMT values for non-zero antibody titers were 145.9 among vaccinated vs 98.4 among unvaccinated.

The potential explanation for these differences is that, in the epidemic season 2021/2022, the dominant virus was influenza A, subtype A/Cambodia/e0826360/2020(H3N2), confirmed by molecular biology methods. The exact subtype was dominant during the 2016/2017 epidemic season, which may be the hypothetic cause of its protective level being the highest, namely, participants may have had contact with the wild virus, which caused immune activation but was too low to cause the infection. Vaccination has not increased the number of people with non-zero titer levels but has increased the titer levels themselves, as observed for GMT values. The A/Victoria/2570/2019 (H1N1)pdm09 subtype seems to be the least effective antigen of the quadrivalent influenza vaccine prepared for epidemic season 2021/2022, as far as reaching the protective level is considered. However, the level of patients with non-zero antibodies has grown significantly. These findings correspond with the research of Kondratiuk et al in the epidemic season 2021/2022. For the group of people older than 65, the highest result was for subtype A/H3N2 and the lowest for A/H1N1/pdm09, as far as antibodies presence is considered. However, among the other 6 age groups (from 0 to 64 years of age) examined, the levels of antibodies varied greatly and were highly heterogeneous [16].

During the epidemic season 2021/2022, COVID-19 continued to pose a threat. Many social movements claimed that vaccinations led to weakened immunity. Our research proved that both the protective levels and GMT values for all influenza subtypes were higher for those vaccinated against COVID-19 than for unvaccinated. As for GMT values, the difference was statistically significant (except of A/H1N1/pdm09), and for protective levels, the difference was clearly visible, especially among both B/lineages and reached 64.2% vs 44.4% (P=0.023) for B/Victoria lineage and 78.6% vs 55.6% (P=0.004) for B/Yamagata lineage. Our results suggest that vaccination against COVID-19 has a positive influence on the immune system and immune response against influenza as well. Those findings align with other those of other researches showing a positive correlation between influenza and COVID-19 vaccinations, most probably due to immune system stimulation [17,18]. They also respond to research conducted in the Influenza Virus Research Institute, National Centre for Influenza at NIZP PZH-PIB, Poland, in the 2019/2020 epidemic season. According to Brydak et al, patients who were vaccinated every epidemic season had milder COVID-19 infections [19].

We also wanted to establish whether patient sex is an important factor in vaccination response. There were significant differences between men and women in GMT values and protective levels for the B/Washington/02/2019(B/Victoria lineage) antigen. Men reached higher values than women (respectively 38.4 vs 67.6; P<0.001, and 58.0% vs 74.6%; P<0.001), which is quite a surprising conclusion, as women in Poland are more eager to vaccinate. Moreover, according to the WHO, post-vaccination antibody titers are higher among older women than men [20]. On the other hand, more current research has shown sex-related differences in humoral immunity after influenza vaccination due to specific genes. According to Yang et al, women respond better to virus A antigens, whereas men respond more to B-line antigens [21].

In our present study, there were no statistically significant differences in protective level and GMT values between younger (65–79 years) and older (≥80 years) seniors. It is rather surprising since, in Poland, the vaccination was free of charge for older seniors, and this group has the highest vaccination rate. According to the meta-analysis published in 2023, systematically repeated influenza vaccination may attenuate effectiveness [22]. However, the non-statistically significant differences in protective level and GMT among patients ≥80 years of age show that their immune response was not better than that among younger, less frequently vaccinated seniors. The possible reason may be age-related involuntary processes of the immune system. To intensify the immunological response, separate vaccines for older people with an increased amount of antigens should be commonly used, as shown in the meta-analysis by Jason et al [23]. In Poland, the standard-dose vaccines (60 μg of hemagglutinin in 0.5 mL) have been used so far, but in the 2024/2025 epidemic season, high-dose vaccines with 240 μg of hemagglutinin in 0.7 mL are expected. Still, the main problem remains a low rate of vaccination. According to the 2009 EU Council Recommendation among older people, 75% vaccination coverage is a goal [24].

Conclusions

In the analyzed group of 256 patients aged 65 to 99 years, different values of GMT and protective levels were recorded depending on the subtype of influenza A and B virus, both B/Yamagata and B/Victoria lines, previous history of vaccination against influenza and COVID-19, and patient sex.

The lowest percentage of patients with antibodies (59.4%), the lowest protective level (27.3%), and the lowest GMT values were recorded for the influenza virus A/Victoria/2570/2019 (H1N1)pdm09 subtype. Only this subtype showed a lack of immune response, in 15 (19.2%) patients, after vaccination. However, vaccination against influenza had almost double the number of patients with non-zero antibody titers for this subtype. Subtype A/Cambodia/e0826360/2020(H3N2) showed the highest protective level (87.5%, 224 patients) and GMT values for non-zero antibody titers, at 145.9 among vaccinated vs 98.4 among unvaccinated, but vaccination did not significantly increase the protective level for this type (86.5% for unvaccinated vs 89.7% for vaccinated; P<0.001). The possible reason is that this subtype was a dominant virus in the 2021/2022 epidemic season. It indicates that the effectiveness of the antigens in the vaccine differs in each epidemic season. High variability and the tendency to constant mutation of the A influenza virus may be a possible reason.

Vaccination against COVID-19 has increased both the GMT values and protective levels for all influenza subtypes. The GMT value difference was statistically significant for A/H3N2 (116.8 vs 71.9; P=0.008), B/Victoria lineage (46.6 vs 27.9; P=0.007), and B/Yamagata lineage (65.9 vs 35.9; P=0.002).

For protective levels, the difference was statistically significant among both B/lineages and reached 64.2% vs 44.4% (P=0.023) for the B/Victoria lineage and 78.6% vs 55.6% (P=0.004) for the B/Yamagata lineage. Post-vaccination immune stimulation may be a reason for this positive correlation.

There were significant differences between men and women in GMT values and protective levels for the B/Washington/02/2019 (B/Victoria lineage) antigen. Men reached higher values than women (respectively 38.4 vs 67.6; P<0.001, and 58.0% vs 74.6%; P<0.001), most probably due to genetic factors.

There were no statistically significant differences in protective level and GMT values between younger (65–79 years) and older (≥80 years) seniors. That indicates that the immune response weakens with aging. In the 2024/2025 epidemic season, a new high-dose quadrivalent influenza vaccine recommended for people 65 years and older will be available in Poland, as older seniors seem to respond less to the standard vaccination.

All of the results show that a new solution, such as preparing different types of vaccine for specific populations, should be considered. Further research and information gathering, as well as popularizing vaccinations globally, are also crucial.