A highly sensitive method for detecting African swine fever virus in oral fluids from naturally infected pigs in Northern Vietnam

A highly sensitive method for detecting African swine fever virus in oral fluids from naturally infected pigs in Northern Vietnam

Thi Ngan Mai^1,4, Thi Huong Giang Tran^1,4, Van Hieu Dong¹, Van Phan Le¹, Thi My Le Huynh¹, Tran Anh Dao Bui¹, Ryoko Uemura², Yasuko Yamazaki³, Thi Lan Nguyen¹ & Wataru Yamazaki³

Early detection and early slaughter through quarantine are essential to prevent the spread of the African swine fever virus (ASFV). Pooled oral fluids testing is a practical approach for pathogen monitoring, but compared to blood, the virus concentration is lower, increasing false negatives. We developed a highly sensitive method for ASFV detection in pig oral fluids using a pretreatment-based concentration protocol. We collected oral fluids from sub-clinical raised pigs in northern Vietnam and conducted a field evaluation using the developed method. A spike test showed up to 100-fold greater sensitivity than a reference method. For performance evaluation, 68 pooled oral fluid samples were collected, of which 63 originated from raised pigs in northern Vietnam and 5 were obtained from healthy pigs in Japan as negative controls. Using real-time PCR, 9/68 (13.2%) were positive by the reference method, while 23/68 (33.8%) were positive by the developed method. Real-time LAMP detected 1/68 (1.5%) and 6/68 (8.8%) respectively. The developed method therefore improved the diagnostic performance and enabled early diagnosis prior to disease onset. These results suggest enhanced sensitivity and feasibility for early ASF diagnosis, potentially contributing to more effective outbreak control. The developed method offers a simple and sensitive tool for rapid ASFV screening in both affected and at-risk farms.

Keywords African swine fever virus, Virion concentration, Early diagnosis, Oral fluid, Pig

Abbreviations

ASF African swine fever

LOD Limit of detection

PBS Phosphate-buffered saline

PEG Polyethylene glycol

SAP Semi-alkaline proteinase

Globalization is accelerating the global spread of transboundary animal diseases (TADs)^1,2. These diseases cause substantial economic losses to the livestock and meat industries worldwide^1,2. Accurate and highly sensitive diagnostics are critical for curbing the transmission of TADs^3,4. However, one persistent bottleneck is the lack of simple and sensitive methods for pathogen detection under field conditions.

African swine fever (ASF) is a highly contagious and lethal hemorrhagic disease affecting domestic and wild suids, caused by ASFV. It is characterized by high fever, hemorrhages in the reticuloendothelial system, and near 100% mortality in susceptible pigs^1,2,5–7. Although initially restricted to sub-Saharan Africa, ASF has become a pandemic across Africa, Eurasia, the Caribbean (Dominican Republic and Haiti), and Oceania (Papua New Guinea), raising concerns about incursions into ASF-free regions such as Japan, the Americas, and Australia.

The 2007 outbreak in Georgia marked the beginning of ASF’s expansion into the Caucasus, Europe, and Russia, demonstrating its transboundary threat^7–9. In 2018, ASF reached China—the world’s largest pig producer—and spread across Southeast Asia, including Vietnam, with devastating impacts on pig farming and food security^1,10–17. As a result, affected countries have suffered significant economic losses^18–20.

Low-virulence, high-transmissibility ASFV genotype I strains emerged in China in 2020²¹, followed by detection of attenuated genotype II strains¹³. In Vietnam, multiple ASFV variants genetically related to Chinese strains are circulating. Recombinant genotype I/II strains were detected in both countries in 2023^14,22. Although Vietnam has authorized vaccines targeting genotype II, emerging recombinant strains may evade immunity¹⁴. Consequently, ASF control remains reliant on biosecurity, rapid diagnosis, and stamping-out policies.

ASFV is shed in various excretions and secretions—including blood, organs, oral fluids, nasal discharges, feces, and urine—through which it can spread^23,24. Oral fluids are widely used for surveillance of pathogens such as porcine circovirus type 2 (PCV2), porcine reproductive and respiratory syndrome virus (PRRSV), swine influenza virus (SIV), and vesicular stomatitis virus (VSV) due to their easy, cost-effective collection^3,25,26. However, lower virus concentrations in oral fluids than in blood contribute to false-negative results^23–25.

Although immunomagnetic separation with virus-specific antibodies provides high sensitivity, this technique is hindered by antibody availability and interference from sample-derived inhibitors^27–29. Previously, we developed a simple and rapid virion concentration method using semi-alkaline protease (SAP) and polyethylene glycol (PEG) for SARS-CoV-2 detection in saliva⁴. This protocol enables viral concentration within one hour. In this study, we adapted the same technique for ASFV detection in pig oral fluids and evaluated its field applicability using 68 pooled samples from pigs in Vietnam and Japan.

Results

A total of 68 pooled oral fluid samples were collected, comprising 63 from raised pigs in northern Vietnam and five from pigs in Japan, which served as negative controls. The diagnostic performance of the developed method was evaluated in comparison with the reference method. The five oral fluid samples from Japanese pigs, used as negative controls, were all negative in both PCR and LAMP assays, as expected (Tables 1, 2; S-Table 1).

Table 1. Diagnostic performance of real-time PCR using the developed and reference DNA extraction methods. Diagnostic sensitivity, 100% (9/9; 95% CI 70.1%–100.0%). Diagnostic specificity, 76.3% (45/59; 95% CI 64.0%–85.3%). PPV, 39.1% (95% CI 22.2%–59.2%). NPV, 100.0% (95% CI 92.1%–100.0%). *Includes five Japanese oral fluid samples used as negative controls. Developed, improved DNA extraction combined with real-time PCR. Reference, column or automated DNA extraction followed by real-time PCR.

Table 2. Diagnostic performance of real-time LAMP using the developed and reference DNA extraction methods. Diagnostic sensitivity, 100% (1/1; 95% CI 20.7%-100.0%). Diagnostic specificity, 92.5% (62/67; 95% CI 83.7%-96.8%). PPV, 16.7% (95% CI 3.0%-56.4%). NPV, 100.0% (95% CI 94.2%-100.0%). *Includes five Japanese oral fluid samples used as negative controls. Developed: Improved DNA extraction combined with real-time LAMP. Reference: Column or automated DNA extraction followed by real-time LAMP.

As presented in Tables 1 and 3 and Fig. 1, all nine samples identified as positive by the reference real-time PCR assay were also detected as positive by the developed method, with lower Ct values as expected due to improved sensitivity. Among the 59 samples that tested negative by the reference method, 45 remained negative using the developed method, while the remaining 14 were newly detected as positive. Accordingly, 9 out of 68 samples (13.2%) were positive using the reference method, whereas 23 samples (33.8%) were positive using the developed method. The use of the developed pretreatment method significantly improved the detection rate of the conventional real-time PCR assay compared to the reference method (p = 0.000512).

The corresponding diagnostic metrics are summarized in Table 1. The real-time PCR assay combined with the developed pretreatment method demonstrated a diagnostic sensitivity and specificity of 100.0% (95% confidence interval [CI] 70.1%–100.0%) and 76.3% (95% CI 64.0%–85.3%), respectively. The positive predictive value (PPV) was 39.1% (95% CI 22.2%–59.2%), and the negative predictive value (NPV) was 100.0% (95% CI 92.1%–100.0%).

For real-time LAMP detection, the developed method identified all one positive sample detected by the reference method and five additional samples that were negative by the reference method, totaling six positives (Table 2 and Fig. 2). The remaining 62 reference-negative samples were also negative by the developed method. Thus, the detection rate increased from 1/68 (1.5%) using the reference method to 6/68 (8.8%) using the developed method. The increase in detection rate achieved by applying the developed viral concentration pretreatment to the conventional real-time LAMP assay, compared to the reference method, was statistically significant (p = 0.0253).

Table 3. C_t values of 23 samples testing positive by the developed method in real-time PCR. C_t, Threshold cycle. No. C_t, No threshold cycle values detected. *10 mL of oral fluid was used for the developed method and 50 μL for the reference method.

As shown in Table 2, the diagnostic sensitivity and specificity of the real-time LAMP assay combined with the developed pretreatment method were 100.0% (95% confidence interval [CI] 20.7%–100.0%) and 92.5% (95% CI 83.7%–96.8%), respectively. The positive predictive value (PPV) was 16.7% (95% CI 3.0%–56.4%), and the negative predictive value (NPV) was 100.0% (95% CI 94.2%–100.0%).

Spiking experiments demonstrated that the developed method achieved up to 1000-fold greater sensitivity than the reference method for real-time PCR detection, and at least 100-fold greater sensitivity for real-time LAMP detection, when pretreatment was performed on 10 mL of oral fluid supernatant within 60 min (Table 4). Results from real-time measurement and visual color change in the LAMP assay were fully consistent (data not shown).

Discussion

We revealed for the first time that a newly developed virion concentration method can reliably detect ASFV in the oral fluids of ASF-infected pigs before the onset of clinical signs, enabling earlier detection. This simple method utilizes large volumes of oral fluid (> 10 ml), resulting in up to 100-fold higher sensitivity than the reference method, which typically uses only 50–200 µl. As > 10 ml of oral fluid is easily obtainable from pig pens, the developed method significantly enhances the diagnostic performance of ASFV surveillance and can strengthen early detection systems worldwide.

The developed method addresses a key limitation of conventional nucleic acid-based diagnostics: false-negative results due to viral loads falling below the limit of detection (LOD). Using the reference method, only 11.1% (1/9) of real-time PCR-positive samples had C_t values below 35.00 (a threshold for definitive positivity), whereas 88.9% (8/9) were borderline (C_t 35.00–36.79). In contrast, with the developed method, 78.3% (18/23) of positive samples had C_t values below 35.00, enabling clear identification of positive cases. Thus, the developed method not only increased the number of positive samples but also shifted more results into the definitive positive range, improving interpretability and diagnostic confidence (Tables 1 and 3, Fig. 1, and S-Table 1).

In the LAMP assay, only one sample tested positive at 39 min 42 s with the reference method, whereas six samples were positive between 23 min 6 s and 39 min 42 s using the developed method (S-Table 1 and Fig. 2). On the other hand, real-time PCR detected 9 and 23 positive samples using the reference and developed methods, respectively. As previously reported³⁰ and demonstrated in spike tests (Table 4), LAMP is intrinsically 10- to 100-fold less sensitive than real-time PCR. This discrepancy, combined with the inherently low ASFV concentration in oral fluids, explains the reduced positivity in LAMP-based detection. Nevertheless, despite its lower sensitivity, the LAMP assay offers significant practical advantages in resource-limited settings. Its rapid reaction time, ease of operation, and independence from advanced laboratory equipment make it highly suitable for routine diagnosis and field deployment, especially in areas lacking real-time PCR infrastructure.

Fig. 1. Comparison of C_t values obtained using the developed and reference methods for 23 oral fluid samples that tested positive by real-time PCR.

ASFV was also detected in oral fluid collected prior to clinical signs and outbreak confirmation (S-Table 1). For instance, at farm B, oral fluid samples collected on July 31 (nos. 9, 12, 14) were positive by the developed method, while pigs showed clinical signs on August 1, and blood tests confirmed ASFV on August 2. The farmer immediately sold all the pigs to the slaughterhouse. These results highlight the value of the developed method for detecting preclinical infections. Of note, these samples were negative by the reference method, further underscoring the improved sensitivity of the developed method.

Despite its simplicity, the multi-step concentration process requires careful handling to avoid cross-contamination. Moreover, some variability in concentration efficiency was observed. For example, samples 51 and 56 showed minimal C_t differences between methods. Possible reasons include ASFV existing as free DNA (detectable only by reference methods) or suboptimal pellet removal during centrifugation, as suggested by preliminary data showing > tenfold loss in recovery when debris was not fully eliminated. Visible pellets were observed in several samples, potentially contributing to these discrepancies.

We also attempted detection in oral fluids from six semi-farmed wild boars (grazing during the day, housed at night), but all were negative (data not shown), likely due to the low ASFV prevalence in asymptomatic wild boars⁹. In future work, we aim to apply this method to wild boar surveillance using bite-rope sampling or other collection techniques.

In this study, pooled oral fluid samples were used to increase group-level detection sensitivity while minimizing cost and invasiveness. Although positive pooled results cannot identify infected individuals, they enable timely countermeasures, including clinical observation, resampling, and containment. This approach aligns with international surveillance priorities for early detection and rapid response.

As limitations, due to resource constraints, we could not perform duplicate testing at the initial analysis stage or confirmatory blood sampling in asymptomatic pigs. In addition, internal positive control (IPC) testing was not conducted due to the same resource limitations. However, oral fluids have demonstrated higher sensitivity for early ASFV detection than blood during the preclinical phase^31,32. The detection of ASFV DNA in clinically healthy pigs on affected farms further supports the diagnostic value of this approach.

We demonstrated the successful detection of ASFV in oral fluids from naturally infected pigs using a method more sensitive than conventional techniques. As oral fluid sampling is already adopted in endemic disease monitoring (e.g., by USDA-APHIS in the United States³³), the developed method has strong potential to enhance existing surveillance systems. Moreover, as a simple, sensitive, and cost-effective method that does not rely on specialized equipment, it may also be adapted for other infectious diseases in livestock, such as foot-and-mouth disease in ruminants and rabies in dogs.

This study highlights the utility of the developed method in enhancing ASFV detection in oral fluids. While the current findings are promising, additional validation is necessary to ensure broader applicability. In particular, inter-laboratory and multi-site evaluations are needed to assess the reproducibility and robustness of the method across different settings and operators. Such future studies will be essential for the wider implementation and standardization of this diagnostic approach.

Fig. 2. Comparison of time-to-positivity (Tp) values obtained using the developed and reference methods for six oral fluid samples that tested positive by real-time LAMP.

Table 4. Limit of detection (LOD) of the developed and reference methods for ASFV detection in spiked ASFV-negative pig oral fluid samples. Ct, Threshold cycle; Tp, Time of positivity; No. Ct, No threshold cycle values detected by real-time PCR; No. Tp, No time-to-positivity values detected by LAMP; ND, Not determined; All results were consistent in duplicate testing, with no discordant outcomes.

Conclusions

We successfully developed a simple and highly sensitive method for concentrating ASFV in the oral fluids of naturally infected pigs and demonstrated that it provides superior detection sensitivity compared to reference methods.

Materials and methods

Ethics approval

All experiments were approved by the Animal Research Ethics Committee (CARE) of the Faculty of Veterinary Medicine, Vietnam National University of Agriculture (Approval No. CARE-2022/05). All procedures were performed in accordance with CARE guidelines and reported following the ARRIVE guidelines (https://arriveguidelines.org).

Epidemiological information and test results for individual pigs

Details are given in S-Table 1 of the supplementary materials.

Oral fluids sampling from raised pigs

As detailed in S-Table 1, a total of 68 pooled oral fluids samples from raised pigs were tested. Of these, 63 samples were collected at an early stage of disease from five farmers (Farms A–E) in four provinces in northern Vietnam between November 20, 2022 and August 25, 2023. On Farms A, D, and E, ASF occurred in one pen and subsequently spread to other pens; samples were collected on different days. On Farms B and C, pigs showed clinical symptoms suspicious for ASF one to two days after sampling. Additionally, five samples were obtained on November 12, 2023, from fattening pigs at five farmers in ASF-free Miyazaki Prefecture, Japan, and used as negative controls (Tables 1, 2 and S-Table 1). Oral fluids were collected using cotton ropes suspended in pens for 30 min²⁶, then transported to the lab. Samples were stored at −80 ℃ (Vietnam) and −30 ℃ (Japan) until use. Prior to testing, samples were thawed, centrifuged at 2000g for 1–5min and supernatants were transferred to 50ml sterile plastic tubes. No more than 75% of the original volume was collected to avoid debris. For limit of detection (LOD) testing, ASFV-negative oral fluids (from four pig farms) were pooled to create at least 100 ml of matrix for spiking.

DNA extraction by reference method

To reduce viscosity and avoid clogging of extraction columns in accordance with the pathogen detection manual 2019-nCoV issued by the National Institute of Infectious Diseases, Japan (NIID-J, 2020)³⁴, 150 μl of oral fluids was diluted 1:3 with 450 μl of phosphate-buffered saline (PBS) in a 1.5 ml-sterile microcentrifuge tube. After vortexing, the diluted oral fluids were centrifuged at 15,600 g for 30 min, 150 μl of the supernatant was extracted using a DNeasy blood & tissue kit (Qiagen, Maryland, USA), and eluted in 50 μl. Alternatively, 200 μl of supernatant was extracted using the mag LEAD 6gC automated extraction platform (Precision System Science Co., Matsudo, Japan) and a reagent cartridge (MagDEA Dx SV, Precision System Science).

ASFV concentration by developed method

A previously described SARS-CoV-2 concentration method4 was adapted for ASFV. Briefly, 10 ml of oral fluid supernatant was transferred to a sterile 50 ml plastic centrifuge tube and diluted 1:3 with 20 ml of SAP (Semi-alkaline proteinase, Suputazyme; Kyokuto Pharmaceutical Industries, Tokyo, Japan). After vortexing, the tubes were incubated at room temperature for 15 min to digest oral fluid components. The mixture was then centrifuged at 15,557 g for 30 min. Subsequently, 24 ml of the supernatant was transferred to a new sterile 50 ml plastic tube and mixed with 9.6 ml of PEG-NaCl solution (equivalent to 40% of the total liquid volume³⁵). The tubes were vortexed and centrifuged again at 8000 g for 20 min to precipitate ASF virions. After the second centrifugation, the supernatant was carefully removed. Then, 100 µl (for use with the DNeasy Blood & Tissue Kit) or 150 µl (for use with the automated extraction platform) was added using a sterile pipette tip, and combined with the approximately 50 µl of residual liquid in the tube, bringing the final volume to approximately 150 μl or 200 μl.

To detach the ASF virion-PEG complex from the tube wall, the pellet (formed during the first centrifugation) was pipetted 10 times at the expected adhesion site using a 1 ml sterile long filtered tip to minimize contamination. A 1 ml sterile short tip was then used to vortex the sample, ensuring complete resuspension of the pellet. The resulting mixture containing the ASF virion–PEG complex was flushed into the tube, and DNA was extracted and purified in a final volume of 50 µl using either a column-based kit or an automated platform, as described above. For four samples (nos. 12, 21–23, shown in S-Table 1) with oral fluid volume was less than 10 ml, the volumes of SAP and PEG-NaCl solutions were adjusted proportionally to maintain the same mixing ratio.

Real-time PCR assay

Real-time PCR was performed on a CFX Opus 96 Real-Time PCR System (Bio-Rad Laboratories, Hercules, CA, U.S.A.) or a QuantStudio 3 (Thermo Fisher Scientific, Inc., Waltham, MA, U.S.A.) with two primers and probes consisting of the forward primer (5′-TGC TCA TGG TAT CAA TCT TAT CG-3′), the reverse primer (5′-CCA CTG GGT TGG TAT TCC TC-3′) and the probe (5′-FAM-TTC CAT CAA AGT TCT GCA GCT CTT-TAMRA-3′) reported by Tignon et al.³⁶. Amplification cycle conditions, reagent volumes and concentrations were slightly modified as follows. Briefly, 50 µl of real-time PCR reaction mixtures were used, comprising 25 µl of 2 × probe qPCR mix (Takara Bio Inc., Otsu, Japan), 0.2 µl each of forward and backward primers (100 µmol l−1 each; Hokkaido System Science, Co., Ltd., Sapporo, Japan), 0.1 µl of probe (100 µmol l−1; Hokkaido System Science), 19.5 µl of nuclease-free water (Takara Bio), and 5 µl of the DNA template. The cycling conditions were as follows: one cycle at 95 ℃ for 30s, followed by 45 cycles each at 95 ℃ for 10 s, and 60 ℃ for 60 s. The automatically calculated Ct value was adopted, and the Ct cut-off value was set at 37.00.

Real-time LAMP assay

Real-time LAMP assay was performed following our previous report30 with slight modifications. Briefly, in-house LAMP reaction mixtures were prepared by doubling the volume of reaction solution to 50 µl with the same composition as described. The amount of template DNA was set to 5 µl. The reaction mixtures were incubated at 63 ℃ for 45 min, followed by 80 ℃ for 5 min to complete the reaction using a real-time turbidimeter (Loopamp EXIA; Teramecs, Kyoto, Japan). When the derivative of turbidity reached 0.05 within 40 min, the reaction was automatically considered positive. Endpoint judgement by unaided eye was also performed. When the reagent color remained purple, the result determined as negative. On the other hand, a change to sky blue was interpreted as positive. For both real-time PCR and LAMP analyses, positive controls were DNA sequences derived from field isolates in Vietnam and artificial synthetic DNA sequences ordered from Eurofins Genomics K. K. (Tokyo, Japan) in Japan.

Preparation of tenfold dilution series of ASFV spiked oral fluids

The ASFV strain ASF/HY01/2019 (GenBank Accession no. MK554698; genotype II) was cultured on porcine alveolar macrophages (TCID₅₀ = 10⁶^.5), stored at −80 ℃, and subsequently diluted tenfold in PBS. The diluted virus stock was then stored at 4 ℃. In parallel, pooled pig oral fluid samples stored at −80 ℃ were thawed at room temperature. The supernatant, created by centrifugation at 2000g for 5min to remove cotton rope and feed residue, was transferred to two new 50m-sterile tubes for the spike test. Next, the pig oral fluid supernatant was dispensed into six 50 ml tubes of 10 ml each, and then, the tenfold dilution series of ASFV was sequentially spiked and vortexed thoroughly. A series of pig oral fluids containing from neat to10⁻5-foldASFV dilutions was prepared, as shown in Table 4.

Determination of LODs

DNA extracted from each dilution using both methods was tested in duplicate by real-time PCR and LAMP. Results were interpreted as positive if both replicates were positive; otherwise, negative (Table 4).

Statistical analysis

Diagnostic sensitivity, diagnostic specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated (Tables 1, 2). Differences between paired proportions were evaluated using McNemar’s test in R (version 4.4.037). Statistical significance was set at p < 0.05 (Figs. 1, 2).

Data availability

The data that supports the findings of this study are available in the supplementary material of this article.

Received: 12 January 2025; Accepted: 15 July 2025

References

1. World Organization for Animal Health [WOAH (formerly known as OIE)]. (2020). Bulletin panorama thematic portfolio. https://oiebulletin.com/wp-content/uploads/2020/Panorama2020-1/panorama-2020-1-en.pdf (accessed 29 May 2025).

2. World Organization for Animal Health [WOAH (formerly known as OIE)]. African swine fever (infection with African swine fever) (Chapter 3.9.1). In Manual of Diagnostic Tests and Vaccines for Terrestrial Animals, 1–18 (2021).

3. Beemer, O. et al. Assessing the value of PCR assays in oral fluid samples for detecting African swine fever, classical swine fever, and foot-and-mouth disease in U.S. swine. PLoS ONE 14(7), e0219532. https://doi.org/10.1371/journal.pone.0219532 (2019).

4. Yamazaki, Y., Alonso, U. A., Galay, R. L. & Yamazaki, W. Development of a simple and highly sensitive virion concentration method to detect SARS-CoV-2 in saliva. Heliyon. 10(12), e33168. https://doi.org/10.1016/j.heliyon.2024.e33168 (2024).

5. Dixon, L. K., Stahl, K., Jori, F., Vial, L. & Pfeiffer, D. U. African swine fever epidemiology and control. Annu. Rev. Anim. Biosci. 8, 221–246.https://doi.org/10.1146/annurev-animal-021419-083741 (2020).

6. Galindo, I. & Alonso, C. African swine fever virus: A review. Viruses. 9(5). https://doi.org/10.3390/v9050103 (2017).

7. Sánchez-Vizcaíno, J. M., Mur, L., Gomez-Villamandos, J. C. & Carrasco, L. An update on the epidemiology and pathology of African swine fever. J. Comp. Pathol. 152(1), 9–21. https://doi.org/10.1016/j.jcpa.2014.09.003 (2015).

8. Rowlands, R. J. et al. African swine fever virus isolate, Georgia, 2007. Emerg. Infect. Dis. 14(12), 1870–1874.https://doi.org/10.3201/eid1412.080591 (2008).

9. Sauter-Louis, C. et al. African swine fever in wild boar in Europe—A review. Viruses https://doi.org/10.3390/v13091717 (2021).

10. Blome, S., Franzke, K. & Beer, M. African swine fever—A review of current knowledge. Virus Res. 287, 198099.https://doi.org/10.1016/j.virusres.2020.198099 (2020).

11. Le, V. P. et al. Outbreak of African swine fever, Vietnam, 2019. Emerg. Infect. Dis. 25(7), 1433–1435.https://doi.org/10.3201/eid2507.190303(2019).

12. Wang, D. et al. Development of a real-time loop-mediated isothermal amplification (LAMP) assay and visual LAMP assay for detection of African swine fever virus (ASFV). J. Virol Methods 276, 113775. https://doi.org/10.1016/j.jviromet.2019.113775 (2020).

13. Zhenzhong, W., Chuanxiang, Q., Shengqiang, G., Jinming, L., Yongxin, H., Xiaoyue, Z., et al. Genetic variation and evolution of attenuated African swine fever virus strain isolated in the field: A review. Virus Res. 319, 198874.https://doi.org/10.1016/j.virusres.2022.198874 (2022).

14. Zhou, X. et al. Emergence of African swine fever in China, 2018. Transbound. Emerg. Dis. 65(6), 1482–1484.https://doi.org/10.1111/tbed.12989 (2018).

15. Zhu, W. et al. Lateral flow assay for the detection of African swine fever virus antibodies using gold nanoparticle-labeled acid-treated p72. Front. Chem. 9, 804981. https://doi.org/10.3389/fchem.2021.804981 (2022).

16. You, S. et al. African swine fever outbreaks in China led to gross domestic product and economic losses. Nat. Food 2(10), 802–808. https://doi.org/10.1038/s43016-021-00362-1 (2021).

17. Zhao, D. et al. Highly lethal genotype I and II recombinant African swine fever viruses detected in pigs. Nat. Commun. 14, 3096. https://doi.org/10.1038/s41467-023-38868-w (2023).

18. Carriquiry, M. E. A., Swenson, D., and Hayes, D. Impacts of African swine fever in Iowa and the United States. Working Paper 20-WP 600, 1–25 (2020).

19. Mason-D’Croz, D. et al. Modelling the global economic consequences of a major African swine fever outbreak in China. Nat. Food. 1(4), 221–228. https://doi.org/10.1038/s43016-020-0057-2 (2020).

20. Nguyen-Thi, T. et al. An assessment of the economic impacts of the 2019 African swine fever outbreaks in Vietnam. Front. Vet. Sci. 8, 686038. https://doi.org/10.3389/fvets.2021.686038 (2021).

21. Sun, E. et al. Genotype I African swine fever viruses emerged in domestic pigs in China and caused chronic infection. Emerg. Microbes Infect. 10, 2183–2193. https://doi.org/10.1080/22221751.2021.1999779 (2021).

22. Le, V. P. et al. Detection of recombinant african swine fever virus strains of p72 genotypes I and II in domestic pigs, Vietnam, 2023. Emerg. Infect Dis. 30(5), 991–994. https://doi.org/10.3201/eid3005.231775 (2024).

23. Davies, K. et al. Survival of African swine fever virus in excretions from pigs experimentally infected with the Georgia 2007/1 Isolate. Transbound. Emerg. Dis. 64(2), 425–431. https://doi.org/10.1111/tbed.12381 (2017).

24. Walczak, M., Żmudzki, J., Mazur-Panasiuk, N., Juszkiewicz, M. & Woźniakowski, G. Analysis of the clinical course of experimental infection with highly pathogenic African swine fever strain, isolated from an outbreak in Poland. Aspects rRelated to the disease suspicion at the farm Level. Pathogens. 9(3), 237. https://doi.org/10.3390/pathogens9030237 (2020).

25. Prickett, J. R. & Zimmerman, J. J. The development of oral fluid-based diagnostics and applications in veterinary medicine. Anim. Health Res. Rev. 11(2), 207–216. https://doi.org/10.1017/S1466252310000010 (2010).

26. Ramirez, A. et al. Efficient surveillance of pig populations using oral fluids. Prev. Vet. Med. 104(3–4), 292–300.https://doi.org/10.1016/j.prevetmed.2011.11.008 (2012).

27. Dhumpa, R. et al. Rapid detection of avian influenza virus in chicken fecal samples by immunomagnetic capture reverse transcriptase-polymerase chain reaction assay. Diagn. Microbiol. Infect. Dis. 69, 258–265. https://doi.org/10.1016/j.diagmicrobio.2010.09.022 (2011).

28. Makino, R. et al. Application of an improved micro-amount of virion enrichment technique (MiVET) for the detection of avian influenza A virus in spiked chicken meat samples. Food Environ. Virol. 12(2), 167–173.https://doi.org/10.1007/s12560-020-09425-1 (2020).

29. Yamazaki, W., Makino, R., Nagao, K., Mekata, H. & Tsukamoto, K. New micro-amount of virion enrichment technique (MiVET) to detect influenza A virus in the duck faeces. Transbound. Emerg. Dis. 66(1), 341–348. https://doi.org/10.1111/tbed.13027 (2019).

30. Ngan MT, Thi My Le H, Xuan Dang V, et al. Development of a highly sensitive point-of-care test for African swine fever that combines EZ-Fast DNA extraction with LAMP detection: Evaluation using naturally infected swine whole blood samples from Vietnam. Vet Med Sci. 2023;9(3):1226-1233. doi:10.1002/vms3.1124

31. Gallardo, C., Soler, A., Rodze, I. & Nieto, R. African swine fever: a global view of the current challenge. Porcine Health Manag. 1,

21. https://doi.org/10.1186/s40813-015-0013-y (2014).

32. Pietschmann, J. et al. Course and transmission characteristics of oral low-dose infection of domestic pigs and European wild boar with a Caucasian African swine fever virus isolate. Arch. Virol. 160(7), 1657–1667. https://doi.org/10.1007/s00705-015-2430-2 (2015).

33. United States Department of Agriculture Animal and Plant Health Inspection Service Veterinary Services. (2022). Swine Hemorrhagic Fevers: African and Classical Swine Fevers. Integrated Surveillance Plan.https://www.aphis.usda.gov/sites/default/files/hemorrhagic-fevers-integrated-surveillance-plan.pdf

34. National Institute of Infectious Diseases, Japan (NIID-J), Pathogen detection manual 2019-nCoV. March 19, 2020 (in Japanese).https://www.niid.go.jp/niid/images/lab-manual/2019-nCoV20200319.pdf (accessed 13 Jan 2024).

35. Yamazaki, Y., Thongchankaew-Seo, U. & Yamazaki, W. Very low likelihood that cultivated oysters are a vehicle for SARS-CoV-2: 2021–2022 seasonal survey at supermarkets in Kyoto, Japan. Heliyon. 8(10), e10864. https://doi.org/10.1016/j.heliyon.2022.e10864 (2022).

36. Tignon, M. et al. Development and inter-laboratory validation study of an improved new real-time PCR assay with internal control for detection and laboratory diagnosis of African swine fever virus. J. Virol. Methods 178, 161–170 (2011).

37. R Core Team. R: A Language and Environment for Statistical Computing (R Foundation for Statistical Computing, 2024).https://www.R-project.org/.

Author contributions

M.T.N.: Data curation, formal analysis, funding acquisition, methodology, resources, visualization, writing—original draft, writing—review and editing. T.T.H.G.: Data curation, methodology, writing—original draft. D.V.H.: Data curation, formal analysis, methodology, visualization. L.V.P.: Resources, methodology. H.T.M.L.: Formal analysis. B.T.A.D.: Formal analysis. R.U: Formal analysis, resources. Y.Y.: Data curation, formal analysis, funding acquisition, methodology, visualization, writing—original draft, writing—review and editing. N.T.L.: Conceptualization, supervision. W.Y.: Conceptualization, data curation, funding acquisition, methodology, pro-ject administration, resources, supervision, validation, visualization, writing—original draft, writing—review and editing.

Funding

This research was supported by Japan Society for the Promotion of Science KAKENHI [Numbers JP21H03180, JP22K05950, JP22KK0097, JP24H00122], Bilateral Program [Number JPJSBP120199944], and the Joint Usage/Research Center for Global Collaborative Research, Center for Southeast Asian Studies, Kyoto University.

Competing interests

The authors ｄｅｃｌａｒｅ no competing interests.

Declaration of submission

The authors confirm that this manuscript or data has not been previously published and is not being considered for publication elsewhere. A preprint has previously been published [Mai et.al 2024]. The authors further confirm that all authors have contributed to the study and have approved the final version.

Additional information

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1038/s41598-025-12139-8.

Correspondence and requests for materials should be addressed to W.Y.

Reprints and permissions information is available at www.nature.com/reprints.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access

This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

© The Author(s) 2025