The use of archival animal samples in this study was reviewed and approved by the Institutional Animal Care and Use Committee of the Faculty of Veterinary Science, Mahidol University (Protocol Number: MUVS-2022-07-44).

The study protocol involving MPXV handling and molecular testing was reviewed and approved by the Institutional Biosafety Committee of Mahidol University (Protocol Number: MU 2022-022).

All laboratory procedures involving primary specimens were conducted within a certified Class II biological safety cabinet in a biosafety level 3 facility. Staff adhered to stringent safety protocols, including the use of double gloves and N95 respirators, and initial sample lysis for DNA extraction was performed to ensure viral inactivation before downstream PCR analysis.

Swab samples from free-ranging M. fascicularis were collected between March and November 2019 (pre-outbreak) across 15 provinces in Thailand, covering the Central (Bangkok, Lopburi, Saraburi, Nakhon Sawan, and Phetchabun), Northeastern (Amnat Charoen, Maha Sarakham, and Mukdahan), Eastern (Chon Buri), Western (Phetchaburi and Ratchaburi), and Southern (Krabi, Narathiwat, Phatthalung, and Satun) regions. Samples from exotic pets and small mammals were collected between May and September 2022 (early-outbreak) in Bangkok, Nakhon Pathom, and Nonthaburi.

For site selection and sampling criteria, provinces were selected for wildlife and exotic animal surveillance based on their status during the 2022–2024 global outbreak to capture a range of exposure risks [16]. Mpox-affected areas were defined as provinces with high human case numbers or significant international travel hubs, including Bangkok (representing approximately 90% of national cases), Nonthaburi, Chon Buri, and a major southern tourist destination such as Krabi [32]. In contrast, non-affected or low-incidence areas were selected across the Northeastern (Amnat Charoen, Maha Sarakham, and Mukdahan), Western (Phetchaburi and Ratchaburi), and inland Central regions (Nakhon Pathom, Lopburi, Saraburi, Nakhon Sawan, and Phetchabun) to provide a comparative baseline of viral absence in animals away from primary transmission hotspots. This dual selection strategy ensured that the ecological baseline covered both high-risk zones for potential human-to-animal spillback and lower-risk rural environments.

In terms of urban and climatic context, the study provinces represented a diverse range of urbanized and natural landscapes within Thailand’s tropical monsoon climate. Urban and peri-urban hubs, including Bangkok, Nonthaburi, Nakhon Pathom, and Chon Buri, are characterized by high human density and fragmented green spaces where wildlife, such as urban rodents and macaques, frequently interact with human populations [38, 39]. In addition, central provinces, including Lopburi and Saraburi, are characterized by significant human–wildlife overlap, particularly with non-human primate populations in urban centers [40]. In contrast, provinces in the Northeastern (Amnat Charoen, Maha Sarakham, and Mukdahan), Western (Phetchaburi and Ratchaburi), Southern (Krabi, Narathiwat, Phatthalung, and Satun), and Central (Nakhon Sawan and Phetchabun) regions consist primarily of rural, agricultural, and mountainous landscapes with higher forest cover, providing distinct ecological settings for potential wildlife reservoirs [41]. While the Central and Eastern regions experience significant temperature fluctuations between the hot and cool seasons, the Southern coastal provinces maintain a more stable, humid tropical climate year-round. Unlike the humid coastal South, the Northeastern provinces experience lower relative humidity and higher peak temperatures during the dry months [42]. This selection accounts for varying ecological factors that influence both the persistence of MPXV in the environment and the risk of spillback into local wildlife reservoirs.

The selection of target animal species was based on specific criteria derived from current scientific literature and ecological risk assessment.

Inclusion criteria were based on two primary ecological and biological factors:

Potential reservoir susceptibility: Species were selected based on evidence from previous mpox outbreaks or experimental studies that identified them as known or suspected reservoirs or susceptible hosts, such as various rodent species and non-human primates.

Based on these criteria, the study focused specifically on free-ranging macaques, small mammals, and exotic pets.

Exclusion criteria included livestock and common domestic animals, such as dogs and cats. These species were excluded from this surveillance effort because they are less likely to serve as primary maintenance reservoirs for MPXV than the targeted wildlife and exotic species during the study period.

Small mammals were identified to the species level primarily through morphological assessment. Field identification followed standard taxonomic keys for Southeast Asian rodents, utilizing external morphometric parameters including body weight, head-and-body length, tail length, condylobasal length, ear length, and hindfoot length [43, 44]. To maintain taxonomic accuracy, immature specimens with ambiguous external features, due to the unreliability of morphometrics in juvenile rodents, were conservatively recorded as unidentified rat species.

Archival swab samples used in this study were categorized into three primary animal groups based on their ecological role and the timing of collection:

Free-ranging macaques (pre-outbreak baseline): A total of 593 M. fascicularis were sampled from 24 free-ranging colonies across 15 provinces between March and November 2019. These pre-outbreak samples were obtained through archival buccal, oral, oropharyngeal, or rectal swabs. Sampling was conducted in collaboration with the Department of National Parks, Wildlife and Plant Conservation (DNP) (Table 1; Supplementary Table S1).

Swab samples were collected by trained veterinarians wearing full personal protective equipment, including gloves, masks, and coveralls. Gloves were changed frequently to prevent cross-contamination. Single-use sterile flocked nylon swabs were used, and each swab was immediately placed into 1 mL of phosphate-buffered saline (pH 7.4) or viral transport medium. Samples were maintained on ice during collection, transported to the laboratory at 4°C within 48 h, and stored at −80°C until use. Freeze–thaw cycles were strictly avoided to maintain sample integrity.

Swab samples were pooled by specimen type to maintain a consistent biological matrix and avoid potential cross-tissue interference. Each individual pool consisted of 4–5 swabs; 40–50 µL was collected from each swab and combined to achieve a final pool volume of 200 µL. This total volume of 200 µL was then used for DNA extraction, ensuring that each individual specimen was adequately represented in the diagnostic reaction.

DNA was extracted from the pooled swabs using the genomic DNA mini kit (Geneaid Biotech Ltd., New Taipei City, Taiwan), according to the manufacturer’s instructions. In total, 553 pooled swabs were processed, with each DNA sample eluted in a final volume of 40 µL.

Target gene selection: To detect the MPXV genome in DNA samples, the B7R gene, which encodes a soluble interferon-gamma receptor, was selected as the diagnostic target because of its high conservation within the MPXV genome [45].

Primer and probe design: To ensure species specificity and prevent cross-reactivity with other orthopoxviruses, such as VARV or vaccinia virus, primers and probes were designed to target unique motifs within the B7R open reading frame [45]. The sequences were as follows:

MPXV_B7R_forward: 5′-ACGTGTTAAACAATGGGTGATG-3′

MPXV_B7R_reverse: 5′-AACATTTCCATGAATCGTAGTCC-3′

MPXV_B7R_probe: 5′-FAM-TGAATGAATGCGATACTGTATGTGTGGG-BHQ1-3′

Reaction condition: Assays were performed in a 20 µL total volume using the QuantiNova Probe RT-PCR Kit (Qiagen, Hilden, Germany). The reaction mixture comprised 5 µL of template DNA, 10 µL of 2× QuantiNova master mix, 0.4 µM of each primer and probe, 1 µL of QN ROX Reference Dye, and 2.8 µL of RNase-free water.

The thermal cycling profile for real-time PCR included initial denaturation at 95°C for 2 minutes, followed by amplification for 40 cycles at 95°C for 15 seconds and 58°C for 60 seconds.

Quality control and contamination prevention: Positive control, consisting of nucleotide design and synthesis from the partial genome of MPXV with 3.7 × 10⁵ copies/µL, and a negative control, consisting of nuclease-free water, were included in each run. To prevent contamination, pre- and post-PCR areas were physically separated, and aerosol-resistant pipette tips were used.

Data analysis: Fluorescence was captured using QuantStudio Software (Applied Biosystems, Waltham, MA, USA).

Result interpretation: A cycle threshold (Ct) value of ≤ 40 was defined as the cutoff for a positive result, consistent with standard diagnostic protocols [46, 47]. This cutoff is widely used in mpox surveillance to maximize sensitivity for low viral load samples. To mitigate the risk of false-positive results at late cycles, any sample with a Ct value of 34 or higher should be re-extracted and re-tested to ensure that no cross-contamination had occurred [48].

Analytical sensitivity: Original validation of B7R-targeted assays using the recombinant plasmid reported a limit of detection (LoD) of approximately 20 copies per reaction [45].

In this study, a total of 1,248 archival animal samples collected across Thailand were tested to detect the presence of MPXV using real-time PCR. To ensure the reliability of potential negative findings, the sample size was evaluated based on its capacity to detect a pre-defined minimum threshold of pathogen prevalence.

In the event of zero positive detections, the 95% confidence interval (CI) for population prevalence was calculated using the rule-of-three (3/n), providing an upper bound for the maximum possible prevalence in the sampled population [49]. Furthermore, surveillance power was assessed as the probability of detecting at least one positive case under various low-level prevalence scenarios, such as 0.1%, 0.5%, and 1% [50]. Calculations were performed using the binomial distribution formula Power = 1 − (1 − d) ⁿ, where d represents the design prevalence and n represents the sample size [51]. For example, the probability that 1,248 samples would have yielded at least one positive result if the virus were actually present at a prevalence of 0.5% can be calculated as Power = 1 − (1 − 0.005) 1248.

Microsoft Excel LTSC Professional Plus 2021 (Microsoft Corporation, Redmond, WA, USA; Version 2108) was used for data management. The prevalence of MPXV was calculated as the proportion of positive results among the total number tested and expressed as a percentage. Data were analyzed using descriptive statistics to estimate the current burden of MPXV and active viral shedding within the sampled population. Inferential statistics were not applied because the cross-sectional design of the study was intended to provide viral prevalence data rather than test specific comparative hypotheses.

The studied population consisted of 29 species representing a total of 1,248 animals, including 593 cynomolgus macaques, 436 exotic pets, and 219 small mammals. The largest proportion of animals (47.5%, 593/1,248) comprised free-ranging M. fascicularis, whose colonies were located close to human habitations and communities in 15 provinces across Thailand (Table 1).

The second largest group consisted of exotic pets (34.9%, 436/1,248) presented to animal hospitals in Bangkok and an adjacent province (Table 2). Among exotic pets, rabbits Oryctolagus cuniculus belonging to the order Lagomorpha represented the majority (56.2%, 245/436), followed by rodents of the order Rodentia (27.5%, 120/436). The remaining animals (16.3%, 71/436) included species belonging to the orders Primates (bushbabies, marmosets, slow lorises, and a tamarin), Carnivora (wild cats, fennec foxes, ferrets, and raccoons), Erinaceomorpha (hedgehogs), and Diprotodontia (sugar gliders and a wallaby).

The final group consisted of small mammals (17.5%, 219/1,248) collected from garbage dumping sites, including rodents of the order Rodentia (89.5%, 196/219), shrews of the order Eulipotyphla (10.0%, 22/219), and a treeshrew of the order Scandentia (0.5%, 1/219) (Table 2). Among these, Rattus tanezumi and Suncus murinus were the most frequently identified species of rats and shrews, respectively. Detailed information on collected samples, including year, location, species, number, and sample types, is summarized in Supplementary Table S1 and Supplementary Table S2.

The distribution of studied animals covered several provinces where human mpox cases had been reported (Figure 1). The majority of animals originated from Central Thailand, particularly Bangkok, which was the main epicenter of the mpox outbreak in the country, with more than 450 cumulative confirmed cases reported [32].

Most exotic pets included in the study were obtained from an animal hospital in Bangkok (68.8%, 300/436), whereas the remaining animals (31.2%, 136/436) were derived from a hospital located in Nakhon Pathom province. Similarly, garbage-scavenging small mammals were collected from dumping site areas in Bangkok (44.7%, 98/219), Nakhon Pathom (9.6%, 21/219), and Nonthaburi (45.7%, 100/219). These two adjacent provinces also reported mpox cases, particularly Nonthaburi, which had the third highest cumulative number of confirmed cases in Thailand (50 cases as of 19 October 2025).

Among free-ranging macaques, nearly one-third of the animals (30.2%, 179/593) were sampled from colonies located in Bangkok and four additional provinces in the central region, including Lopburi, Saraburi, Nakhon Sawan, and Phetchabun, all of which reported small numbers of mpox cases. Additional macaque populations were sampled from the Northeastern region (12.4%, 74/593), eastern region (16.9%, 100/593), western region (20.6%, 122/593), and southern region (19.9%, 118/593).

Some macaque colonies were located in provinces affected by the outbreak, including Maha Sarakham (northeast), Chon Buri (east), Phetchaburi (west), and Krabi (south). Among these, Chon Buri reported the second highest cumulative number of confirmed cases in Thailand (85 cases as of 19 October 2025), whereas the remaining provinces reported at least one confirmed case since the outbreak began.

DNA extracted from swab samples of all 1,248 animals was tested for MPXV using real-time PCR targeting the B7R gene. All samples were negative for MPXV DNA, with the cutoff for a positive result defined as Ct ≤ 40 cycles. No inconclusive or borderline Ct values were observed. All samples produced a “not detected” result, and no late-cycle amplification signal below the threshold was recorded.

These findings indicate that MPXV DNA was not detected in any of the sampled animals at the time of collection.

All 1,248 animals tested negative for MPXV DNA by PCR. Based on this sample size, the upper limit of the 95% CI calculated using the rule-of-three indicated that the prevalence of MPXV in the studied population was <0.24%.

Power analysis demonstrated that a sample size of 1,248 provided a 99.81% probability of detecting at least one positive case if the true prevalence of MPXV in the population was 0.5%. The study therefore achieved the standard 95% confidence threshold for excluding a prevalence of 0.24% or higher.

In addition, statistical power remained above 70% for detecting a prevalence as low as 0.1%, indicating that the absence of positive results represents a statistically robust lack of widespread viral circulation in the studied animal populations.

The current study examined the presence of MPXV DNA in archival swab samples from wild animals and exotic pets collected in 2019 and 2022. A total of 1,248 swab samples derived from 593 cynomolgus macaques, 436 exotic pets, and 219 small mammals were found negative for MPXV DNA. Although no viral DNA was identified from the animals during the study period, our finding provides baseline evidence of MPXV absence in wildlife and exotic animal populations in Thailand prior to and during the early phase of the global 2022 outbreak. Evidence of absence demonstrated through well-designed studies is critical for risk assessment, surveillance prioritization, and policy decision-making, particularly in the context of emerging zoonotic threats.

The novelty of this study lies in the use of pre- and early-outbreak archival samples, multi-population risk-based animal selection, and focus on spillback prevention in a non-endemic country. To our knowledge, this is the first nationwide molecular surveillance study in Southeast Asia assessing MPXV absence in wildlife, exotic pets, and garbage-scavenging mammals during the 2022 global outbreak. A model of One Health surveillance framework that integrates wildlife, exotic animals, and environmental exposure pathways demonstrated by this study can be adapted by other non-endemic countries during emerging zoonotic outbreaks.

Historically, Thailand has no record of human and animal cases of mpox. However, the recent largest outbreak in 2022 affected more than 120 countries worldwide, of which at least 115 countries are located outside endemic regions where mpox had never been previously reported, including Thailand [16, 52]. The ongoing global outbreak caused more than 163,000 cases with more than 420 reported deaths as of 30 September 2025 [16]. Thailand reported the first human case in July 2022 in a Nigerian man who had entered Thailand since October 2021 [53]. From the first case identified, the Thai population has been infected with MPXV, resulting in more than 930 confirmed cases and 13 deaths so far [16, 32]. Similar to other non-endemic countries, the MPXV outbreak in Thailand has been classified as a subclade IIb strain. However, the first case of subclade Ib was confirmed in Thailand in August 2024, in which the case had traveled from an African country [54]. No additional case infected with subclade Ib had been reported in Thailand until now.

Based on genome sequence data, the 2022 outbreak has clearly occurred through the exportation of MPXV from affected areas of Africa to other countries [55, 56]. The virus has been spread mainly through sexual or intimate contact [10, 11, 52]. However, the cryptic transmission in human and animal hosts that probably occurred prior to 2022 in a non-endemic country should be taken into consideration [55, 57]. The inclusion of archival animal samples collected prior to the 2022 global outbreak represents a rare opportunity to assess cryptic circulation of MPXV before widespread human transmission in a non-endemic region. Additionally, a wide variety of animal species can become infected with MPXV via direct contact or fomites [58]. Infected animals, especially wildlife, can become new reservoirs of MPXV in unaffected geographic regions [34, 58]. Although the specific reservoir species of the virus have not been identified, several potential animal reservoirs, such as rodents and non-human primates, may harbor the virus, leading to occasional spillover events to humans [55, 57]. Thus, monitoring and surveillance of animals at high-risk of viral infection, particularly those in close contact with infected persons, are necessary to detect and possibly prevent the establishment of animal reservoirs [34].

A convergence of extensive urban wildlife populations, widespread exotic pet ownership, and close human–animal contact would make a non-endemic country like Thailand a high-risk interface for the establishment of MPXV and an ideal sentinel setting for early spillback surveillance. We attempted to include potential animal populations for MPXV detection in the pre- and early-2022 mpox outbreak. This study uniquely integrates three distinct animal populations representing direct contact, trade-associated, and environmentally exposed pathways, enabling a comprehensive assessment of MPXV spillback risk across the human–animal–environment continuum.

Free-ranging M. fascicularis were included in the study because non-human primates are susceptible to MPXV infection. Moreover, Thailand has a serious problem of macaque overpopulation in several cities, especially tourist destinations [59]. All macaque populations in the study lived close to human settlements, placing them at high-risk of virus transmission through direct human contact.

Another target population for MPXV detection was exotic pets, in which all exotic animals in this study were owned pets and several species can be infected with MPXV. Exotic pets are considered wild species rather than domesticated species [60]. Moreover, exotic animals are often non-native or non-indigenous, and some are imported from other countries [61]. Without proper pathogen screening, imported animals can facilitate the transmission of MPXV, and the global trade of these animals creates environments where close, cross-species contact significantly amplifies viral transmission and spreads infectious diseases across borders. According to evidence from the 2003 outbreak in the U. S., MPXV was transmitted through imported rodents from Africa, in which prairie dogs and several native species became infected following close contact with those imported animals [29]. In this study, the exotic pets, although lacking a history of exposure to human mpox cases, were nonetheless considered a potential source of virus infection and transmission.

The last population studied for the presence of MPXV comprised small mammals scavenging around garbage dumping sites. In urban areas of Thailand, landfills are the most common form of waste disposal. Household waste is collected and transported to landfill or garbage dumping sites for disposal. Most landfill sites are open areas that cannot limit animal access, resulting in potential risk of virus transmission to animal scavengers [62]. Generally, poxviruses can persist from weeks to several months, depending on surface materials and environmental conditions [33, 63–65]. A recent study demonstrated that vaccinia virus was stable on different surfaces such as plastic, towel, paper, and glass, and the virus could persist much longer at low temperature at 4°C and −20°C [66]. Exposure to contaminated environments can potentially transmit infectious viruses. Hence, garbage-scavenging animals could be at high-risk of contracting disease through environmental exposure, particularly to domestic waste originating from settings where mpox cases have been reported. In this study, small mammals representing garbage-scavenging species were sampled from garbage dumping sites in Bangkok and two additional provinces with confirmed mpox cases. However, no link was identified between the animal collection sites and the human case locations or contaminated waste.

In the group of non-human primates, M. fascicularis, bushbabies (Galago sp.), marmosets (Callithrix sp.), slow lorises (Nycticebus sp.), and tamarin (Saguinus midas) were included in our study, in which M. fascicularis were free-ranging animals while the others were represented in the group of exotic pets. Based on animal experiments, M. fascicularis has been found to be highly susceptible to MPXV infection through intravenous exposure [67, 68]. The infected macaques showed clinical signs resembling human mpox, including vesiculopapular rash, fibrinonecrotic pneumonia, lymphadenopathy, and death within 9–17 days post-infection [67, 68]. M. fascicularis exposed intranasally or via aerosol to MPXV have also exhibited symptoms associated with mpox [69, 70]. Marmosets Callithrix sp. were another non-human primate that has been used as an animal model for mpox study, but their clinical signs resembled smallpox rather than mpox [71].

Potential reservoirs among rodent species were also investigated for MPXV in the current study. Thirteen different rodent species were included, in which 10 species of rodents (squirrels, prairie dogs, dumbo rats, bamboo rats, woodland dormouse, guinea pigs, chinchillas, and hamsters) were represented in the group of exotic pets and four species of rats (Pacific rats, Norway rats, Oriental house rats, and unidentified rats) were represented in the group of garbage-scavenging animals. All rodent species can possibly be infected with MPXV, especially prairie dog Cynomys sp., which was known to be a potential animal reservoir for MPXV [72]. The 2003 outbreak investigation in the U. S. demonstrated that black-tailed prairie dogs Cynomys ludovicianus that had contact with infected African rodent species were accidentally infected, became sick, and died during the outbreak [31]. Experimentally infected prairie dogs have exhibited clinical characteristics resembling human mpox disease. Earlier studies in the prairie dog model have confirmed high levels of MPXV shedding via oral, nasal, ocular, and rectal routes, as well as animal-to-animal oral, respiratory, and mucosal transmission [73–75]. Therefore, this species has been widely used as a model to study the pathogenesis and transmission of MPXV [76, 77]. Another susceptible species for MPXV infection is an African dormouse species Graphiurus sp., in which Graphiurus murinus was present in our exotic pet population. Previous studies of mpox have used Kellen’s dormouse Graphiurus kelleni as an animal model. Kellen’s dormouse has been shown to be highly susceptible to intranasal infection with clade I MPXV, and infected animals could shed the virus in nasal washes, suggesting that this species can shed infective virus into the environment via nasal secretions [78].

Other mammal species in our studied population, such as hedgehogs, rabbits, and shrews, are known to be susceptible to MPXV infection. Hedgehogs and rabbits were represented in the exotic pet population, while shrews were represented in the garbage-scavenging animal population in our study. MPXV infection under natural conditions has been evident in hedgehogs and shrews. Following the 2003 MPXV outbreak in the U. S., hedgehogs Atelerix sp. sharing housing areas close to infected African rodents tested positive for MPXV. Further investigation confirmed the presence of viral DNA in various tissue samples of the infected hedgehogs [29]. On the other hand, antibodies against OPXV have been detected in shrews from Zambia and the DRC at rates of 33.3% (14/42) and 1.2% (1/84), respectively [79, 80]. For rabbits, this species has been tested for MPXV infection in the laboratory. Susceptibility depended on the inoculation method and the animals’ age [81]. Infected rabbits exhibited acute illness with a generalized rash when challenged via the intravenous route, but not via oral inoculation [81]. Moreover, young rabbits have been shown to be more susceptible to MPXV infection than adult rabbits [81]. However, no evidence of natural MPXV infection in rabbits has been reported.

Sample pooling is a widely recognized strategy for large-scale surveillance, particularly when expected pathogen prevalence is low (typically <5%–10%). In this study, pooling 4–5 swabs per reaction was utilized to maximize laboratory throughput and optimize reagent consumption. While pooling increases population-level sensitivity by allowing the screening of a larger number of animals, it introduces a potential reduction in individual diagnostic sensitivity due to the dilution effect. Previous studies have shown that pooling five samples typically increases the Ct value by approximately 2.0 to 2.6 cycles [82]. Consequently, the risk of false-negative results is highest for specimens with low viral loads (e.g., Ct > 34), as the diluted target may fall below the assay’s LoD [83, 84]. However, for small pool sizes (≤ 5), sensitivity for moderate-to-high viral loads remains high (90%–100%), whereas sensitivity for low viral loads may decrease significantly [85].

Although real-time PCR results from this study indicated an absence of MPXV viral DNA in all tested animals, these negative results should be interpreted with caution. Potential false negatives could have resulted from suboptimal sample collection timing, the sample pooling effect, or poor sample quality. While PCR provides definitive evidence of active infection, it offers only a temporal snapshot of viral presence within the sampled animals. Our reliance on molecular detection means that prior infectious shedding or past exposure within these populations cannot be ruled out. Furthermore, as this study focused exclusively on PCR detection of MPXV DNA, we did not assess antibody status, viral infectivity, or transmission potential among sampled animals. To address this critical gap in understanding long-term viral circulation, particularly for archival samples collected in 2019, future research should prioritize serosurveillance. Implementing antibody-based screening alongside molecular detection is essential to map the full extent of past exposures and identify potential silent reservoirs in non-endemic regions.

Based on our findings and the documented risks, we propose actionable recommendations for a robust One Health surveillance framework in non-endemic regions by focusing on three aspects:

Integration with human surveillance: Establish formal data-sharing protocols between public health and veterinary health agencies to facilitate rapid temporal and spatial correlation of emerging cases in both populations [86].

Pet trade regulation: Implement mandatory health screening and certification programs for all imported exotic mammals, particularly rodents and non-human primates known to be susceptible to orthopoxviruses, to prevent disease introduction via trade routes [87].

Wildlife–urban interface monitoring: Develop standardized, long-term monitoring programs for key sentinel species (e.g., macaques and urban-dwelling small mammals) in high-contact areas, using both molecular and serological methods to track viral circulation baselines and detect emergence events [88].

This study has several limitations regarding the sampled animal populations and the diagnostic methodology employed. First, the low number per animal species limited virus detection and may not reflect the true prevalence of the disease in individual animal species. Furthermore, the study lacked molecular species confirmation (e.g., DNA barcoding of cytB or COI genes) for a subset of the sampled rodents, particularly those classified as unidentified rats. While morphological keys are standard for field studies, they may fail to distinguish between cryptic species common in Southeast Asia [89, 90]. This limits our ability to make definitive ecological inferences regarding species-specific reservoir competence, as different species within the same genus may vary in their capacity to maintain or transmit MPXV.

Second, the samples used in this study were opportunistically repurposed, which may affect the validity of the results. In addition, the period of sample collection and the habitation areas of the studied animals may not have been entirely relevant to the presence of human cases in the country. Nevertheless, samples collected prior to the outbreak could provide baseline information on the virus in potential animal reservoirs, particularly in a non-endemic region. Although this study demonstrated geographic overlap between animal sampling sites and human mpox cases, we could not establish a precise temporal linkage between animal sampling and human epidemiological trends. The pre-outbreak samples were collected in 2019, prior to the first laboratory-confirmed human mpox case in Thailand; therefore, no temporal overlap existed for this cohort to evaluate potential spillover or spillback events. Regarding the early-outbreak samples collected in 2022, while this period coincided with the emergence of the global outbreak, high-resolution temporal and spatial data for human cases in the specific sampling provinces were not publicly available to allow direct correlation. Consequently, we could not determine whether animal sampling exactly preceded or followed local peaks in human cases, which limits our ability to definitively assess the immediate risk of anthropozoonotic spillback during that period.

Third, single-time-point sampling of an individual hindered the ability to detect the virus during active infection in animal reservoirs because of the rapid clearance of the virus from mucosal surfaces and major organ systems [80, 91]. This is consistent with the previous report that viral shedding could be observed only briefly, between 5 and 22 days, in the African rope squirrel model [92]. Hence, repeated sampling is required to rule out infection in animal reservoirs.

Fourth, sample pooling may lead to a loss of diagnostic sensitivity due to dilution of low-titer viral DNA. Pooling dilution may result in underestimation of viral presence when individual viral loads are near the detection limit [83, 84].

Fifth, in the absence of serological testing, this study could not identify prior exposures or infections in the population. Because our methodology relied exclusively on PCR, which detects active viral shedding, we cannot rule out the possibility that MPXV circulated in these populations prior to sampling. This is particularly relevant for the archival pre-outbreak samples collected in 2019; animals may have cleared the virus before sampling, leaving no detectable DNA despite a history of infection.

Lastly, the lack of environmental sampling, such as testing wastewater, bedding, or soil/fomites at scavenging sites, may have obscured broader indicators of viral circulation within the study area. Environmental surveillance is a critical tool for detecting pathogens in a community before individual clinical cases emerge [93–96]. Future research should prioritize these methods within a One Health framework to better understand viral persistence and transmission dynamics in these settings.