Effectiveness of a trivalent inactivated split-virion influenza vaccine (Tri Fluvac) in Thailand: a phase IV test-negative case-control study Authors: Wachiraporn Arunothong, Thundon Ngamprasertchai, Janjira Thaipadungpanit, Napat Phetkub, Amornkarn Thongbai, Kanokwan Saengsawang, Ponthip Wirachwong, Isariya Techatanawat, Somchaiya Surichan, Piengthong Narakorn, Kritchawat Boonya-anan, Amnat Khamsiriwatchara, Jaranit Kaewkungwal, Saranath Lawpoolsri Published: 2026 ABSTRACT SummaryBackgroundThailand has invested in domestic influenza vaccine manufacturing to strengthen health security, yet real-world evidence on the effectiveness of locally produced vaccines is limited. This study aimed to evaluate the effectiveness of a domestically produced trivalent inactivated split-virion influenza vaccine (GPO Tri Fluvac) against medically attended outpatient laboratory-confirmed influenza infection following its introduction into the population.MethodsWe conducted a phase IV test-negative case–control study among adults (≥18 years) presenting with influenza-like illness (ILI) in Lampang Province, Thailand, during the 2024–2025 influenza season. Laboratory-confirmed influenza was identified by RT-PCR. Vaccination status was obtained from hospital information system records. Vaccine effectiveness (VE) was estimated using multivariable logistic regression. Influenza virus clades and subclades were identified through viral genome sequencing. The study was registered in the Thai Clinical Trials Registry (TCTR20240909002).FindingsAmong 4975 participants with ILI, 703 (14%) had laboratory-confirmed influenza. GPO Tri Fluvac vaccination was received by 103 (15%) of 703 influenza cases and 1611 (38%) of 4272 test-negative controls. The adjusted VE of GPO Tri Fluvac against medically attended outpatient laboratory-confirmed influenza was 73.1% (95% CI: 66.3–78.5). Subgroup VE estimates by age group and virus subtype were presented descriptively. VE against hospitalization could not be reliably estimated because of the small number of severe outcomes. Genome sequencing showed that most circulating viruses belonged to the same major clades as the corresponding Southern Hemisphere 2024 vaccine strains included in GPO Tri Fluvac.InterpretationGPO Tri Fluvac provided substantial real-world protection against laboratory-confirmed influenza among adults in Thailand. These findings support its use in the national influenza vaccination program and highlight the value of domestic vaccine production for regional health security.FundingThailand Science Research and Innovation through the National Vaccine Institute and the Government Pharmaceutical Organization. Research in contextEvidence before this studySeasonal influenza vaccination reduces morbidity and mortality, but vaccine effectiveness (VE) varies by season, circulating strains, and population. To assess existing regional literature, a PubMed search was conducted for peer-reviewed articles published up to June 30, 2026, using the search term (“influenza vaccine” OR “flu vaccine”) AND (“vaccine effectiveness” OR “vaccine efficacy”) AND (“Southeast Asia”). Evidence from Southeast Asia remains limited, particularly for domestically manufactured vaccines. Most available VE estimates are derived from sentinel surveillance and mixed-vaccine settings, with limited data evaluating the real-world performance of newly introduced, domestically manufactured vaccines.Added value of this studyThis study is the first, to our knowledge, to assess real-world VE of a domestically produced trivalent inactivated influenza vaccine (GPO Tri Fluvac) in Thailand using a test-negative case–control design with standardized ILI criteria and RT-PCR confirmation. We demonstrate substantial protection against medically attended outpatient influenza among adults aged ≥18 years, with comparable effectiveness across age groups and virus subtypes. Viral genomic sequencing showed alignment with the 2024 Southern Hemisphere (SH2024) vaccine strains, supporting the observed VE.Implications of all the available evidenceThese findings provide critical post-licensure evidence supporting the use of locally manufactured influenza vaccines in national immunization programs. Demonstrating real-world effectiveness of domestic vaccine production strengthens pandemic preparedness, regional health security, and sustainable vaccine self-reliance in Southeast Asia. Research in context Evidence before this study Seasonal influenza vaccination reduces morbidity and mortality, but vaccine effectiveness (VE) varies by season, circulating strains, and population. To assess existing regional literature, a PubMed search was conducted for peer-reviewed articles published up to June 30, 2026, using the search term (“influenza vaccine” OR “flu vaccine”) AND (“vaccine effectiveness” OR “vaccine efficacy”) AND (“Southeast Asia”). Evidence from Southeast Asia remains limited, particularly for domestically manufactured vaccines. Most available VE estimates are derived from sentinel surveillance and mixed-vaccine settings, with limited data evaluating the real-world performance of newly introduced, domestically manufactured vaccines. Added value of this study This study is the first, to our knowledge, to assess real-world VE of a domestically produced trivalent inactivated influenza vaccine (GPO Tri Fluvac) in Thailand using a test-negative case–control design with standardized ILI criteria and RT-PCR confirmation. We demonstrate substantial protection against medically attended outpatient influenza among adults aged ≥18 years, with comparable effectiveness across age groups and virus subtypes. Viral genomic sequencing showed alignment with the 2024 Southern Hemisphere (SH2024) vaccine strains, supporting the observed VE. Implications of all the available evidence These findings provide critical post-licensure evidence supporting the use of locally manufactured influenza vaccines in national immunization programs. Demonstrating real-world effectiveness of domestic vaccine production strengthens pandemic preparedness, regional health security, and sustainable vaccine self-reliance in Southeast Asia. Introduction Influenza is a vaccine-preventable disease for which seasonal vaccines have been used for more than six decades.1 Nevertheless, influenza causes seasonal epidemics worldwide. Influenza virus types A and B are predominantly circulating and causing seasonal influenza epidemics.2 The Southeast Asia Region (SEAR) is considered a hotspot for influenza emergence due to its high population density, socioeconomic constraints, and relatively low vaccination coverage.3 Thailand has a well-established seasonal influenza vaccination policy, recommending vaccination for high-risk groups such as children <2 years, adults ≥65 years, individuals with chronic diseases, pregnant women, healthcare workers, and specific occupational groups including Hajj/Umrah pilgrims and poultry cullers.3 Nevertheless, despite the national program, Thailand reported more than 600,000 influenza cases and 51 associated deaths in 2024.4 To strengthen national and regional health security, Thailand incorporated the establishment of domestic influenza vaccine production into its National Strategy for Pandemic Influenza Preparedness in 2005. The Government Pharmaceutical Organization (GPO) was funded to develop industrial-scale vaccine manufacturing capacity for pandemic and seasonal influenza vaccines. With technical support from the World Health Organization (WHO), an egg-based, inactivated subunit influenza vaccine was successfully developed and demonstrated acceptable safety and immunogenicity in both preclinical animal and clinical studies.5,6 In March 2024, the GPO Tri Fluvac vaccine was approved by the Food and Drug Administration, Ministry of Public Health, Thailand. It is a split virion, inactivated influenza vaccine containing A/Victoria/4897/2022 (H1N1) pdm09-like virus; A/Thailand/8/2022 (H3N2)-like virus; and B/Austria/1359417/2021-like virus (B/Victoria lineage), consistent with WHO recommendations of influenza virus vaccine for use in the 2024 southern hemisphere influenza season.7 Assessment of vaccine effectiveness (VE) following the introduction of influenza vaccines into routine immunization programs is critical to determine their performance under real-world conditions.1 In contrast to pre-licensure clinical trials, post-marketing VE studies reflect the complex interplay between circulating viral strains, host characteristics, health-care–seeking behavior, and implementation contexts.8 Evidence generated from such evaluations is essential for quantifying the population-level impact of vaccination, informing national immunization policies, guiding future vaccine composition and deployment strategies, and sustaining confidence in influenza vaccination programs.1,3 In this context, this study aimed to assess the real-world vaccine effectiveness (VE) of the trivalent inactivated split-virion influenza vaccine (GPO Tri Fluvac) against medically attended influenza illness in the 2024–2025 season. The findings provide essential post-licensure evidence to support national influenza immunization policy following the introduction of a domestically manufactured trivalent influenza vaccine. Methods This study employed an unmatched test-negative case–control design, a WHO-recommended approach for estimating influenza vaccine effectiveness (VE).1 The test-negative design could potentially reduce selection bias of cases and controls due to difference in health seeking behavior.9,10 A single-catchment-area, multi-site study, was conducted in Muang District, Lampang Province, Thailand. Participants with influenza-like illness (ILI) were identified and enrolled through routine outpatient services at Lampang Hospital, two community health centers, and 13 sub-district health-promoting hospitals under the supervision of Lampang Hospital, covering a population of approximately 100,000. During August 2024, before the start of enrollment, approximately 23,000 doses of GPO Tri Fluvac were allocated and administered in the district as part of the public seasonal influenza vaccination program. Participant enrollment began after completion of the GPO Tri Fluvac vaccination campaign. Although other influenza vaccines were provided to specific high-risk groups through the national program, GPO Tri Fluvac was administered to adults in the general population as well as eligible high-risk groups. Vaccination records indicated that approximately 36% of GPO Tri Fluvac recipients were aged ≥60 years. The target population comprised adult residents (≥18 years) living within the Lampang Hospital catchment area. The eligibility criteria for ILI were those presenting within 10 days of symptom onset with acute respiratory infection defined by measured temperature ≥37.8 °C (or history of fever with antipyretic use within 6 h), and cough and/or sore throat, following the ILI definition by the Centre for Disease Control and Prevention (CDC).11 The exclusion criteria included onset of ILI during hospitalization (nosocomial infection); inability to verify influenza vaccination history within the previous year; receipt of any influenza vaccine other than GPO Tri Fluvac since January 2024; receipt of antiviral agents for the current illness episode; and receipt of GPO Tri Fluvac within 14 days before symptom onset. Patients attending the outpatient departments of participating facilities first underwent routine general screening according to each facility's standard outpatient workflow. Adult patients identified as having acute fever and who expressed interest in participating in the study were referred to the research team for study information, informed consent, and assessment of eligibility criteria. Eligible patients attending participating facilities were consecutively screened and recruited between 19 September 2024 and 31 May 2025. Individuals could be re-enrolled for subsequent ILI episodes within the study period, retaining the same study identifier. All participants were enrolled at outpatient presentations and underwent nasopharyngeal swab collection for laboratory-confirmed RT-PCR for influenza virus. Appropriate treatment was provided following the hospital standard treatment guidelines for ILI. Hospitalization due to severe acute respiratory infection (SARI) after enrollment was assessed during follow-up by telephone on day 7 (+3 days) and verified using hospital information system records. Thus, SARI was evaluated as a post-enrollment severity outcome among participants enrolled with ILI, rather than as a separate inpatient enrollment category. History of receipt of the 2024 seasonal influenza vaccine was obtained at enrollment through participant self-report and review of vaccination documents, when available. Vaccination status was further verified during analysis using records from the hospital information system (HIS), which captures all vaccinations administered under the national seasonal influenza vaccination program. Nasopharyngeal swab samples were tested for influenza virus and subtype by RT-PCR at the certified laboratory of Lampang Hospital. RT-PCR–positive samples underwent whole-genome sequencing at the Faculty of Tropical Medicine, Mahidol University, using Oxford Nanopore Technology (ONT). Influenza virus HA clades and subclades were assigned using Nextclade,12 and maximum-likelihood phylogenetic trees were constructed to compare study viruses with reference sequences obtained from the GISAID database.13 Statistical analysis Baseline demographic and clinical characteristics were summarized and compared between influenza cases and test-negative controls using appropriate parametric or non-parametric tests for continuous variables and chi-square or Fisher's exact tests for categorical variables. Vaccine effectiveness (VE) against laboratory-confirmed influenza among ILI populations was estimated using logistic regression. The odds ratio (OR) was defined as the odds of GPO Tri Fluvac vaccination among participants with laboratory-confirmed influenza compared with that among test-negative controls. VE was calculated as 1 minus odds ratio (OR). Potential confounders were identified using directed acyclic graphs (DAGs) to map the assumed causal relationships between influenza vaccination and laboratory-confirmed influenza (Supplementary Figure S1). Based on the DAG, age, sex, BMI, underlying medical conditions in the past year, previous influenza vaccination, and previous influenza infection were considered potential confounders; however, previous influenza vaccination and previous influenza infection were not measured and could not be included in the model. Healthcare-seeking behavior was not adjusted for because, in the test-negative case–control design, it was considered part of the case-ascertainment process and a potential collider. The final multivariable logistic regression model was adjusted for age, sex, BMI, and underlying medical conditions in the past year. Calendar time was not included because vaccination was administered before enrollment through a time-limited vaccination program, making time-dependent vaccination during follow-up unlikely to introduce substantial bias.14, 15, 16 Adjusted ORs and 95% confidence intervals were then estimated from the final model. The primary analyses for VE were based on HIS-verified vaccination status and individual influenza outcome, defined as having at least one ILI episode with influenza virus infection throughout the study period. Self-reported vaccination history was additionally evaluated in sensitivity analyses to assess potential information bias. To account for repeated illness episodes, an additional episode-level sensitivity analysis was performed using generalized linear mixed models, with participant identifier included as a random effect to account for within-person correlation. Subgroup VE analyses were performed by age group (18–59 years, ≥60 years) and influenza subtype (A/H1N1, A/H3N2, and B), as prespecified in the study protocol. These subgroups were selected based on clinical and immunological relevance, as vaccine immune responses and influenza risk may differ between adults and older adults, and vaccine-induced protection may vary by influenza subtype. These analyses were intended to provide descriptive VE estimates within prespecified clinically relevant subgroups, rather than to statistically compare VE estimates between subgroups. The genetic relatedness of RT-PCR–positive influenza viruses were described using phylogenetic analysis and compared with that of vaccine strains. A prespecified statistical analysis plan was finalized before database lock. All analyses were conducted using SAS software, Version 9.4. (SAS Institute Inc., Cary, NC, USA). Sample size calculations followed WHO guidance for test-negative VE studies.17 Assuming an anticipated VE of 50%, vaccine coverage of 25%, a case-to-control ratio of 1:4, and the lower bound of two-sided 95% confidence interval above 40%,18 a minimum of 925 cases and 3700 controls was required. To account for attrition, approximately 5000 ILI patients were targeted for enrollment. Adverse events related to study procedures (e.g., nasopharyngeal swab) were reported in accordance with Good Clinical Practice guidelines. Vaccine-related adverse events were monitored through routine post-marketing pharmacovigilance and were not directly assessed in this study. Ethical statement The study was approved by the Ethics Committee of the Faculty of Tropical Medicine, Mahidol University (MUTM 2024-044-01 and MUTM 2024-044-02), and the Ethics Committee of Lampang Hospital (EC 081/67). All study procedures complied with the Declaration of Helsinki and applicable ethical standards and were performed following written informed consent from all participants. Role of the funding source This study was funded by the Thailand Science Research and Innovation (TSRI) through the National Vaccine Institution (NVI) and the Government Pharmaceutical Organization (GPO). The GPO, sponsor, was responsible for ensuring that the study was conducted in accordance with the International Council for Harmonization Good Clinical Practice (ICH-GCP) guidelines, including quality assurance and quality control procedures. External clinical monitoring was provided by the sponsor to ensure protocol adherence and data integrity. Data management was conducted independently by the centre of excellence for biomedical and public health informatics (BIOPHICS), Mahidol University. The principal investigator retained primary responsibility for the conduct of the study and oversight of data analysis, interpretation, and manuscript preparation, without influence from the funding source on the study conclusions. The study was registered in the Thai Clinical Trials Registry (TCTR20240909002). Results Participant characteristics A total of 4999 participants were screened and provided written informed consent, contributing to 5088 influenza-like illness (ILI) episodes. Of these, 23 participants (24 episodes) did not meet the eligibility criteria, and a nasopharyngeal swab was not collected for one participant. Consequently, 4975 participants comprising 5063 ILI episodes were included in the ILI analysis population. Among these, 703 (14%) participants had RT-PCR–confirmed influenza infection. Hospitalization due to SARI after outpatient ILI enrollment occurred in 76 participants, of whom 19 had RT-PCR–confirmed influenza infection (Fig. 1). Monthly participant enrollment is shown in Fig. 2. Enrollment of influenza-like illness (ILI) cases was highest between January and March, with a secondary increase in May. In contrast, the peak of laboratory-confirmed influenza cases occurred in January and February 2025.Fig. 1Consort diagram. N = number of participants; NE = number of episodes; ILI = Influenza-like Illness; SARI population = Subsequent Severe Acute Respiratory Infection among participants enrolled with ILI; PCR = RT-PCR for influenza virus.Fig. 2Monthly distribution of test-positive cases and test-negative controls during the study period. Enrollment duration: 19 September 2024–31 May 2025. Consort diagram. N = number of participants; NE = number of episodes; ILI = Influenza-like Illness; SARI population = Subsequent Severe Acute Respiratory Infection among participants enrolled with ILI; PCR = RT-PCR for influenza virus. Monthly distribution of test-positive cases and test-negative controls during the study period. Enrollment duration: 19 September 2024–31 May 2025. In the ILI population, cases and controls differed by sex, BMI, and prevalence of underlying medical conditions (Table 1). Cases were more frequently male than controls (41% vs 32%; p < 0.0001), had a higher median BMI (24.2 vs 23.1 kg/m2; p < 0.0001), and more often reported at least one underlying medical condition in the preceding year (30% vs 25%; p = 0.006). Age distribution and median age was not statistically significantly different between groups.Table 1Characteristics of participants with influenza-like illness (ILI) and severe acute respiratory infection (SARI), by influenza infection status (case and control).ILI (N = 4975)p-valueSARI (N = 76)p-valueCase (N = 703)Control (N = 4272)Case (N = 19)Control (N = 57)Sex; N (%) Male289 (41)1363 (32)<0.0016 (32)22 (39)0.58 Female414 (59)2909 (68)13 (68)35 (61)Age group (years); N (%) 18–59550 (78)3236 (76)0.153 (16)14 (25)0.54 ≥ 60153 (22)1036 (24)16 (84)43 (75)Age (years) Median (q1–q3)39.0 (28.0–58.0)44.0 (22.0–59.0)0.6070.0 (64.0–83.0)69.0 (60.0–75.0)0.13 Min-Max18–9518–9526–9521–89BMI (kg/m2) Median (q1–q3)24.2 (21.3–27.6)23.1 (20.6–26.0)<0.00122.4 (19.0–27.8)20.4 (17.8–23.3)0.26 Min-Max13.3–46.712.4–79.013.3–31.313.7–45.7Medical condition during the past year; N (%) Yes210 (30)1068 (25)0.00616 (84)44 (77)0.75 No493 (70)3203 (75)3 (16)13 (23)Influenza vaccination; N (%) GPO Tri Fluvac103 (15)1611 (38)<0.0010 (0)3 (5)0.98 Not vaccinated600 (85)2653 (62)19 (100)54 (95) Other influenza vaccinea0800a8 participants who received other influenza vaccines, as recorded in the hospital information system, were excluded from vaccination effectiveness analysis. Table 1Characteristics of participants with influenza-like illness (ILI) and severe acute respiratory infection (SARI), by influenza infection status (case and control).ILI (N = 4975)p-valueSARI (N = 76)p-valueCase (N = 703)Control (N = 4272)Case (N = 19)Control (N = 57)Sex; N (%) Male289 (41)1363 (32)<0.0016 (32)22 (39)0.58 Female414 (59)2909 (68)13 (68)35 (61)Age group (years); N (%) 18–59550 (78)3236 (76)0.153 (16)14 (25)0.54 ≥ 60153 (22)1036 (24)16 (84)43 (75)Age (years) Median (q1–q3)39.0 (28.0–58.0)44.0 (22.0–59.0)0.6070.0 (64.0–83.0)69.0 (60.0–75.0)0.13 Min-Max18–9518–9526–9521–89BMI (kg/m2) Median (q1–q3)24.2 (21.3–27.6)23.1 (20.6–26.0)<0.00122.4 (19.0–27.8)20.4 (17.8–23.3)0.26 Min-Max13.3–46.712.4–79.013.3–31.313.7–45.7Medical condition during the past year; N (%) Yes210 (30)1068 (25)0.00616 (84)44 (77)0.75 No493 (70)3203 (75)3 (16)13 (23)Influenza vaccination; N (%) GPO Tri Fluvac103 (15)1611 (38)<0.0010 (0)3 (5)0.98 Not vaccinated600 (85)2653 (62)19 (100)54 (95) Other influenza vaccinea0800a8 participants who received other influenza vaccines, as recorded in the hospital information system, were excluded from vaccination effectiveness analysis. Characteristics of participants with influenza-like illness (ILI) and severe acute respiratory infection (SARI), by influenza infection status (case and control). 8 participants who received other influenza vaccines, as recorded in the hospital information system, were excluded from vaccination effectiveness analysis. Within the SARI population no statistically significant differences were observed between influenza cases and test-negative controls with respect to sex, age, BMI, or underlying medical conditions (Table 1). Influenza vaccination history Overall, 1714 participants (35%) received GPO Tri Fluvac. After verification using hospital information system (HIS) records, eight participants with documented receipt of other influenza vaccines were excluded from the vaccine effectiveness analysis. Misclassification between self-reported vaccination status and HIS records was most frequently observed among older adults and participants with severe acute respiratory infection (SARI) (Supplementary Table S1). Among 703 influenza positive cases, 103 (15%) had received GPO Tri Fluvac, compared with 1611 (38%) of 4654 test-negative controls; none of the influenza-infected SARI cases had received GPO Tri Fluvac (Table 1). Vaccine effectiveness, ILI population Vaccine effectiveness (VE) of GPO Tri Fluvac against laboratory-confirmed influenza among medically attended outpatient ILI cases is presented in Fig. 3. The adjusted VE in the overall ILI population was 73.06% (95% CI: 66.25–78.9). In age-stratified analyses, VE was 72.34% (95% CI: 63.82–78.86) among adults aged 18–59 years and 74.61% (95% CI: 61.28–83.35) among adults aged ≥60 years. Subtype-specific analyses showed protection against all major circulating influenza subtypes, with the VE against influenza A(H1N1) (80.18%, 95% CI: 67.88–87.77), influenza B (70.40%, 95% CI: 61.13–77.45), and influenza A(H3N2) (70.89%, 95% CI: 47.62–83.82).Fig. 3Vaccine effectiveness in preventing laboratory-confirmed influenza, in terms of medically attended outpatient influenza-like illness, by age groups and influenza virus subtypes. Vaccine effectiveness in preventing laboratory-confirmed influenza, in terms of medically attended outpatient influenza-like illness, by age groups and influenza virus subtypes. Vaccine effectiveness, SARI population Among participants with severe acute respiratory infection (SARI), no laboratory-confirmed influenza cases occurred among those who received GPO Tri Fluvac (Table 2). Given the small number of SARI outcomes, vaccine effectiveness against SARI could not be estimated with sufficient precision and is therefore not reported.Table 2Distribution of SARI participants by laboratory-confirmed influenza infection and GPO Tri Fluvac vaccination status.Influenza positive case (N)Influenza test-negative control (N)TotalGPO Tri FluvacTotalGPO Tri FluvacTotal190543Age groups 18–59 years30141 ≥60 years160432Influenza virus strains H1N190573 H3N260573 B40573 Table 2Distribution of SARI participants by laboratory-confirmed influenza infection and GPO Tri Fluvac vaccination status.Influenza positive case (N)Influenza test-negative control (N)TotalGPO Tri FluvacTotalGPO Tri FluvacTotal190543Age groups 18–59 years30141 ≥60 years160432Influenza virus strains H1N190573 H3N260573 B40573 Distribution of SARI participants by laboratory-confirmed influenza infection and GPO Tri Fluvac vaccination status. Sensitivity analysis Sensitivity analyses using alternative definitions of vaccination status (self-report vs HIS records) and analytic units (individual vs ILI episode) produced VE estimates consistent with the primary analysis, supporting the robustness of the findings (Supplementary Figure S2). Circulating influenza virus Among the 703 RT-PCR–confirmed influenza cases, genome sequencing was successfully performed in 232 nasopharyngeal samples. As shown in Table 3, circulating influenza viruses were classified by clade and subclade and compared with the corresponding Southern Hemisphere 2024 vaccine strains included in the GPO Tri Fluvac formulation. Most sequenced influenza A(H1N1), A(H3N2), and B viruses belonged to the same major clades as the corresponding vaccine strains, although some were classified into different subclades. These genetic data were used to describe circulating virus lineages during the study period and were not interpreted as direct evidence of antigenic similarity.Table 3Influenza type A and B virus circulating during the study period (N = 232).influenza virus clade/subcladeReceived GPO Tri FluvacYes N (%)No N (%)Total34 (100)198 (100) H1: 5a.2a.1/D.3.12 (5.9)14 (7.1) H1: 5a.2a/C.1.95 (14.7)31 (15.7) H1: 5a.2a/C.1.9.101 (0.5) H1: 5a.2a/C.1.9.303 (1.5) H3: 2a.3a.1/J.22 (5.9)7 (3.5) H3: 2a.3a.1/J.2.225 (73.5)108 (54.5) H3: 2a.3a.1/J.2.401 (0.5) B: 3a.2/C.5.101 (0.5) B: 3a.2/C.5.6024 (12.1) B: 3a.2/C.5.6.101 (0.5) B: 3a.2/C.5.707 (3.5)GPO Tri Fluvac SH2024 vaccine strain: H1N1: 5a.2a.1/D; H3N2:2a.3a.1/J; and B:3a.2/C. Table 3Influenza type A and B virus circulating during the study period (N = 232).influenza virus clade/subcladeReceived GPO Tri FluvacYes N (%)No N (%)Total34 (100)198 (100) H1: 5a.2a.1/D.3.12 (5.9)14 (7.1) H1: 5a.2a/C.1.95 (14.7)31 (15.7) H1: 5a.2a/C.1.9.101 (0.5) H1: 5a.2a/C.1.9.303 (1.5) H3: 2a.3a.1/J.22 (5.9)7 (3.5) H3: 2a.3a.1/J.2.225 (73.5)108 (54.5) H3: 2a.3a.1/J.2.401 (0.5) B: 3a.2/C.5.101 (0.5) B: 3a.2/C.5.6024 (12.1) B: 3a.2/C.5.6.101 (0.5) B: 3a.2/C.5.707 (3.5)GPO Tri Fluvac SH2024 vaccine strain: H1N1: 5a.2a.1/D; H3N2:2a.3a.1/J; and B:3a.2/C. Influenza type A and B virus circulating during the study period (N = 232). GPO Tri Fluvac SH2024 vaccine strain: H1N1: 5a.2a.1/D; H3N2:2a.3a.1/J; and B:3a.2/C. Discussion This study provides post-licensure evidence on the real-world effectiveness of the domestically produced trivalent inactivated split-virion influenza vaccine (GPO Tri Fluvac) in Thailand. Using a test-negative case–control design, we demonstrated substantial protection against laboratory-confirmed influenza among medically attended outpatient cases, with consistent effectiveness across age groups and circulating virus subtypes. These findings indicate that the vaccine performs effectively under routine programmatic conditions and supports its use within Thailand's national influenza prevention strategy. The vaccine effectiveness observed in this study was 73% (66%–78%) in preventing medically attended outpatient influenza infection among adult ≥18 years. Influenza VE is known to vary across seasons, vaccine formulations, circulating strains, and study populations.19 The VE observed here is relatively higher than estimates reported during the 2024/2025 season in the United States and Europe, where VE ranged from approximately 30% to 60%.20,21 In prior seasons, VE estimates from Italy and Ireland ranged between 40% and 60%, while studies from Thailand and China during the 2022/2023 season reported VE of 51% and 48%, respectively.22, 23, 24, 25, 26 A pooled VE estimate for the 2023 Southern Hemisphere influenza vaccine from Australia, South Africa, and Thailand was 65% (95% CI: 56–76%), comparable to our findings.27 Differences across studies may partly reflect methodological variation. Unlike surveillance-based studies relying on heterogeneous ILI definitions and mixed vaccine exposures, this study evaluated a single vaccine (GPO Tri Fluvac) using a standardized ILI case definition, which may have contributed to more precise VE estimate observed. Seasonal influenza vaccination has been associated with protection against severe acute respiratory infection (SARI) requiring hospitalization. A pooled VE estimate from eight countries including Thailand, reported a VE of 52% (37%–67%) against influenza-associated SARI during 2023 season, with comparable estimates observed in the United States and Germany during 2024 season.17,18,28 In addition, a multicounty evaluation from five South American countries reported an adjusted VE of 35% against influenza-associated hospitalization among high-risk groups, using vaccines based on the WHO-recommended 2024 Southern Hemisphere formulation.29 In this study, no laboratory-confirmed influenza-related SARI cases were identified among recipients of GPO Tri Fluvac. However, the number of SARI outcomes was insufficient to reliably estimate VE against severe influenza-related illnesses. Similar constraints related to sample size and statistical uncertainty have been reported in prior evaluations of influenza VE against severe outcomes,25 underscoring the need for adequately powered studies with targeted enrollment in inpatient settings to better characterize vaccine protection against hospitalization. Older adults are at higher risk of influenza-related morbidity due to increased susceptibility and reduced vaccine immunogenicity.30 Although immunogenicity studies and clinical trials have shown diminished immune responses in this age group, high-dose influenza vaccines—licensed for use in older adults—have demonstrated enhanced immunogenicity, though further evidence on real-world VE remains needed.31 Real-world studies of standard dose influenza vaccine have generally reported comparable VE between adults and older adults.20,21,27 Consistent with this evidence, our study estimated VE of 75% among older adults and 73% among adults aged 18–59 years. Vaccine effectiveness is influenced by population-level vaccine coverage.19 In this setting, older adults were prioritized for influenza vaccination through government-supported programs, including GPO Tri Fluvac and other influenza vaccines, which may have contributed to the observed protection. Subtype-specific analyses demonstrated substantial effectiveness against influenza A(H1N1) and influenza B. Genome sequencing showed that most circulating A(H1N1), A(H3N2), and B viruses belonged to the same major clades as the corresponding SH2024 vaccine strains included in GPO Tri Fluvac.7 However, these subtype-specific VE estimates should be interpreted descriptively because this study was not designed to formally compare VE across subtypes, and genetic sequencing alone cannot determine antigenic match without antigenic assays such as hemagglutination inhibition or microneutralization assays. These results support the value of continued virological monitoring and antigenic characterization to inform timely vaccine strain selection. This study provides real-world evidence on the effectiveness of a domestically produced seasonal influenza vaccine. The test-negative case–control design minimized selection bias and confounding related to health care seeking behavior. The study was conducted in a well-defined geographic area with low population mobility, and ILI enrollment was restricted to the vaccine catchment area, reducing exposure misclassification and enhancing internal validity. The enrollment period covers both 2024 Southern and Northern Hemisphere influenza season, in which the recommended composition of influenza virus vaccines was not changed. Vaccination history misclassification was minimized by using hospital information system records as the primary data source, with sensitivity analyses confirming the robustness of the findings. Influenza-like illness was defined using standardized criteria, and outcomes were based on laboratory-confirmed influenza by RT-PCR, ensuring validity. In addition, viral genome sequencing enabled characterization of circulating influenza viruses by clade and subclade relative to the vaccine components, providing supportive context for the interpretation of the VE estimates. This study has several limitations. Its single-province design may limit generalizability to settings with different health-care utilization patterns, vaccine coverage, and influenza transmission dynamics; however, the temporal patterns of ILI and influenza enrollment in this study were consistent with national surveillance data.4 Recruitment was limited to public-sector health facilities and may not fully capture individuals seeking care in private settings, although recipients of government-supported influenza vaccination are more likely to access public services. Despite adjustment for key covariates, residual confounding from unmeasured factors, such as prior influenza infection, influenza vaccination history, or social contact patterns, cannot be excluded. Co-circulating vaccine-preventable respiratory infections, such as SARS-CoV-2, may also bias VE estimates in test-negative studies; however, a sensitivity analysis excluding SARS-CoV-2-positive participants produced a similar VE estimate, suggesting minimal influence on the primary findings. Finally, the small number of severe outcomes limited the precision of VE estimates against SARI, underscoring the need for larger, multi-site inpatient setting studies to better evaluate protection against severe influenza. In conclusion, this study provides post-licensure evidence that a domestically produced trivalent influenza vaccine confers substantial real-world protection against laboratory-confirmed influenza among adults in routine clinical practice. GPO Tri Fluvac demonstrated substantial protection against laboratory-confirmed influenza among adults in routine clinical settings, with comparable effectiveness across age groups, virus subtypes, and meaningful protection against circulating virus strains. These findings emphasize the strategic value of domestic influenza vaccine manufacturing for strengthening national and regional health security, ensuring timely vaccine availability, and sustaining pandemic preparedness. Ongoing vaccine effectiveness monitoring and integrated virological surveillance will remain essential to guide vaccine composition, optimize deployment strategies, and inform evidence-based influenza immunization policy. Contributors SL, WA, TN, and JK contributed to study conceptualization and methodology. SL, TN, JT, and AK undertook data curation, formal analysis, and data visualization. SL, PW, IT, SS, and PN secured funding. SL, WA, TN, JT, NP, and AT conducted the investigation and site operation. KB and AK were responsible for project administration. JT performed genomic sequencing. WA, NP, and KS conducted and validated laboratory diagnostics. SL, PW, PN, and JK provided supervision. SL, WA, TN, and AK performed data validation. SL drafted the initial version of the manuscript. SL and TN had access to raw data and verified the data. SL and WA had final responsibility for the decision to submit for publication. All authors contributed to data interpretation, critically revised the manuscript, and approved the final version. Data sharing statement De-identified individual participant data will be available upon reasonable request to the corresponding author, following approval by the study investigators and in accordance with applicable ethical and regulatory requirements. Declaration of interests SL is the principal investigator who received funding support from the Government Pharmaceutical Organization (GPO). PW, IT, SS, PN, and KB are employees of the Government Pharmaceutical Organization (GPO), which manufactures of the GPO Tri Fluvac influenza vaccine evaluated in this study. All other authors declare no competing interests.