AFRICAN SWINE FEVER: A BIOSECURITY CHALLENGE FOR DOMESTIC PIG PRODUCTION IN SERBIA

African swine fever (ASF) is currently the most important challenge for domestic pig production worldwide. Th e virus reached Eurasia in 2007, and is today aff ecting more than half of European Union member countries. Among Western Balkan countries, Serbia suff ered the fi rst case of ASF in a backyard holding in 2019. Since then, numerous outbreaks in domestic pigs and wild boar have been reported throughout the country despite the eff orts of the veterinary authorities to control the disease. Th e lack of an eff ective vaccine is one of the main constraints, and the only currently available option to prevent further ASF infections is the application of strict biosecurity measures. Regarding biosecurity, backyard pig producers and smallholding farmers in Serbia have substantial gaps in the knowledge and fail to comply with safe production behaviour that favours the spread of ASF virus. In the currently prevailing smallholder and backyards farming systems, farm biosecurity is largely non-existent. Th e aim of this review was to identify specifi c ASF-risks factors in the current pig production system and gaps in biosecurity measures related to the human activities recognised as social and cultural identity in Serbia. Moreover, the main risk factors for ASF spreading and transmission at the domestic/wild boar interface, biosecurity practices in diff erent production systems, and possible future control measures and awareness campaigns are discussed.


INTRODUCTION
African swine fever (ASF) is a viral disease of domestic pigs and wild boars that currently represents a major threat to the swine industry worldwide  Liu et al., 2021). Within European continent, the fi rst ASF incursion was reported in Portugal in 1957 (Boinas et al., 2011). It took until 1995 to offi cially eradicate the disease from the continent, with the exception of Sardinia Island, where ASF is still endemic from 1978 to this day (Cappaia et al., 2018). In 2007, a new introduction of the highly virulent genotype II of African swine fever virus (ASFV) was reported in Georgia, the Caucasus (Rowlands et al., 2008). Despite all preventive measures, in early 2014, ASF reached the territory of European Union (EU) with the fi rst case reported in Lithuania followed by other Baltic countries (Blome et al., 2020;Malakauskas et al., 2022). Since then, disease has continuously been spreading, and it reached Asia (2018) and the Americas (2021), which identifi ed ASF as the worst livestock pandemic of this century (Liu et Ardelean et al., 2021). Indeed, in this part of Europe, the observed epidemiological ASFV pattern is diff erent and mostly associated with the characteristics of the domestic pig sector with large percentage of backyards with low levels of biosecurity where pig keeping represents an important source of livelihood and a traditional heritage (Chenais et al., 2019b;Gervasi and Guberti, 2021).
Since at present there is no treatment or vaccination, prevention and control of ASF rely on biosecurity measures (Jurado al., 2018;Bellini et al., 2021). It is well known that ASF virus can be transmitted to domestic pigs and wild boar through direct and/or indirect contact with infected animals, contaminated fomites and through the ingestion of contaminated meat products (Olesen et al., 2020;Bellini et al., 2021). Th e virus is highly resistant to environmental conditions and can persist in contaminated fomites and meat products for several months, contributing to the disease spread (Mazur-Panasiuk et al., 2019; Beato et al., 2022). In many cases, the exact route of introduction into domestic pig herds cannot be determined, but most introductions are attributed to indirect virus transmission (Ståhl et al., 2019;Olesen et al., 2020). However, in the current epidemic involving domestic pigs and wild boar population in Europe, disease transmission is frequently driven by human activities (EFSA, 2019). Indeed, the specifi c infl uence of humans and social, cultural factors on ASF epidemiology today is increasingly recognized (Jurado al., 2018;EFSA, 2022). It is important to mention that ASFV is the only known DNA virus that can be transmitted by vectors Since then, eight Ornithodoros species have been found to be involved in the ASFV transmission (Blome et al., 2020). However, a new type of ASFV that can infect hard ticks and transmitted from female adults to the fi rst generation larvae was discovered in China. However, the conducted studies did not confi rm that hard ticks are able to transmit ASFV to susceptible pigs (Liu et al., 2021). Other insects that may mechanically spread virus have also been reported, for example, the stable fl ies. In the study by Olesen et al. (2018), it was demonstrated that blood-feeding fl ies are capable of transporting infectious virus for at least 12 hours. Th erefore, fl ies may play a role in the introduction of ASFV to pig farms, even those with high biosecurity ( In this review, we explored the pig production systems in Serbia and identifi ed specifi c ASF-challenge risks in biosecurity measures related to human activities recognised as social and cultural identity. Also, the main risk factors of disease spreading and transmission at the domestic/wild boar interface, the importance of biosecurity in diff erent production systems, and control eff orts that require further attention in awareness campaigns are discussed.

OVERVIEW OF ASF SITUATION IN SERBIA (2019-2021)
In 2019, ASF was for the fi rst time detected in domestic swine backyard population in central part of Serbia (Milićević et al., 2019). Following the fi rst one, several outbreaks were confi rmed in the villages of one municipality. Disease outbreaks were registered in the next 2 months in the backyard domestic pig population. According to the epidemiological investigation, almost all cases were fi rstly detected as health disorders in sows (anorexia, abortion, death) . However, the last outbreak in 2019 was confi rmed in another region, at the border with Romania in the south-east area of Vojvodina province . Th e distance between the fi rst and the last outbreaks was estimated to about 185-190 km. According to the results of epidemiological investigations, it was concluded that these outbreaks represent most probably two separate introductions from a neighbouring country -Romania . In Europe, the long distance ASF transmission has been associated with the disposal of infected waste, meat or meat products in wild boar habitat, for example, in the Czech Republic, where the closest ASF cases were about 400-500 km away (Jurado et al., 2018;Bellini et al., 2021). Similarly, in 2017, in Romania, the fi rst detection of ASF was confi rmed in a domestic pig backyard, and contaminated Ukrainian meat products were suspected as the most likely source of ASFV (Zani et al., 2019). Subsequently, the Romanian domestic pig sector was aff ected with more than 3800 outbreaks from 2017 to 2020 (Ardelean et al., 2021). According to the results of epidemiological investigation in Serbia, it has been suggested that ASFV was most probably brought in with the diff erent meat products from the aff ected

THE ORGANISATIONAL STRUCTURE OF DOMESTIC PIG PRODUCTION SECTOR IN SERBIA
In the Western Balkan region, the existence of highly variable pig farming system was reported (Prodanov-Radulović et al., 2015; EFSA, 2019). Considering the situation in Serbia, the main and highly important diff erence as compared to the EU member countries is related to the structure and organisation of domestic pig sector (Prodanov-Radulović et al., 2020; Prodanov-Radulović et al., 2020b). Serbia has the highest pig density of all countries in the Balkan region, with a total population of 2.7 million pigs (EFSA, 2019). However, the estimated number of pigs has fl uctuated over the years, and nowadays it shows constant decreasing. Based on the offi cial data, there were 2,983,102 pigs in 2020, and from that number 1,260,970 were in Vojvodina Province . Th e offi cial data of the Veterinary Directorate diff er from the above due to the fact that only marked individuals are recorded in the database. However, in terms of percentage, over 50% of the domestic pig population is located in production units with very low or no biosecurity measures (Prodanov-Radulović et al., 2020a; Polaček et al., 2021). According to the offi cial data, more than 40% of the people live in rural areas, and 36% of the total population is poor or at risk for poverty (EFSA, 2019). Th e above data indicate that this could be the actual percentage of human population oriented towards extensive pig production sector. Serbia has started to harmonise the legislation related to animal diseases with the EU, but it is still ranked among the most vulnerable for disease outbreaks in Europe (EFSA, 2019; EFSA, 2022).
In the EU, pig farms are classifi ed into three categories: non-commercial farms (pigs kept only for fattening for own consumption and neither pigs nor any of their products leave the holding); commercial farms (sell the pigs or move pig products off the holding) and outdoor farms (pigs kept temporarily or permanently outdoors) . Rather than taking into account the farm size, this classifi cation considers the commercial attitude of the holdings. In this way, it controls two important facts: the risk of ASF spreading by trading pigs and the risk for the farm of being exposed to source of infection (EFSA, 2019; EFSA, 2022). However, the pig farming in Serbia include fi ve different pig production holdings: commercial pig farm (industrial pig production); family farm of type A (farm with more than 10 animals and high level of biosecurity); family farm of type B (farm with more than 10 animals and low level of biosecurity); backyards (few animals/less than 10 pigs kept mainly for self-consumption, with a low or total absence of biosecurity) (Prodanov-

BIOSECURITY CHALLENGE IN COMMERCIAL PIG PRODUCTION SECTOR IN SERBIA
Despite the fact that all the mentioned diff erent pig production units are highly diverse, according to the Serbian Law, they are in the category of commercial holdings. Actually, anyone who has a registered pig holding (regardless of number/type of production) and carries out the offi cial pig labelling can obtain the necessary veterinary health certifi cate and sell pigs on the country market (Prodanov-Radulović et al., 2015; Prodanov-Radulović et al., 2020a). Of course, the actual commercial production implies intensive pig production from farrow-to-fi nish and/or farrow-to-piglets or fattening only. Lately, the commercial fattening units can be found quite frequently in the villages, where weaned pigs imported from ASF-free EU member countries are placed (Prodanov-Radulović et al., 2017a; Prodanov-Radulović et al., 2020b). Within the actual pig unit classifi cation, there are substantial diff erences in only one classifi ed group, i.e., in the commercial pig holdings. In general, the true commercial pig farms include the classical pig farm structure according to the technological pig production process. Again, there is a diff erence between old types of pig holdings (built in the period 1970-1980s) and farms built aft er 2000, where the breeding pigs imported from EU are mainly located. Th e number of these farms is actually small, and they are usually a part of the regional companies (Prodanov-Radulović et al., 2020b). On the other hand, there are commercial pig holdings, which in were earlier owned by the state (socialist model of governance). In the 1990s, farms were privatized but the new owners have not been obligated to invest in the modern concept of biosecurity. Indeed, today we have examples of a large production system, with capacity from 450 to 1300 sows that is still managed by the old type of farrow-to-fi nish production system. Some of the biosecurity measures that we know today, which are essential for sustainable pig production, are not possible to be implement without major investments In backyards and smallholders, pig feeding strategies depend on feed resources availability and the ability of farmers to buy ingredients. In these systems, domestic pigs frequently have access to swill (i.e., the kitchen left over food from owners and restaurants) (Chenais et al., 2019a; Mutua and Dione, 2021). According to the law, swill feeding is banned in Serbia; however, it is diffi cult to control in practice. Similarly, swill feeding was banned in the EU in 2002, but the epidemiological studies of ASF outbreaks have shown that this practice is still used (Boklund et al., 2020). Such practices can facilitate ASFV spread, as happened in 2012, when ASF was introduced in Ukraine due to the use of contaminated pork products in swill feeding (EFSA, 2019; . In general, swill is considered to be the most likely source of ASF introduction in Georgia (de la Torre  Th e existence of breeding animals in the backyards (gilts, sows, boars) is highly-risky situation in the extensive production system. Moreover, in Ser-bia, breeding boars are frequently found in the small-scale holdings, and they are moved around for breeding purposes (Prodanov- Radulović et al., 2015;Polaček et al., 2021). Th e critical point from this aspect is animal loan practices, i.e., sharing of boars for natural mating in several villages (Mutua and Dione, 2021). Th is can be signifi cant risk factor contributing to the transmission of ASFV through direct pig-to-pig contact (Olesen et al., 2020). Th e aforementioned specifi c circumstances are likely to contribute to the introduction and establishment of ASFV in Serbian vulnerable pig production system thus promoting the disease spreading. On the contrary, according to the EU regulations, sows or boars cannot be held on non-commercial farms for mating purposes (Cappaia et al., 2018;Bellini et al., 2021).
Another important risky activity in backyards and smallholdings is related to home-slaughtering practice. Backyard pigs are mostly slaughtered at home, usually before Christmas or whenever new meat supplies for family are needed . Home-slaughtering is considered as a feature of non-professional pig production, which is a known constraint to ASF control (EFSA, 2019; Bellini et al., 2021). Such practices contribute to the spread of ASFV due to improper disposal of off al, oft en in the immediate environment of the village, and the use of slaughter waste for feeding other domestic animals in the yard (dogs, cats). Finally, home-slaughtering could be a driver for the spread of the other infectious and zoonotic diseases (Petrović et al., 2019;Petrović et al., 2022). However, home slaughtering of domestic pigs is allowed in Serbia.
It is well-known that implementation of biosecurity is a key to successful pig production in an ASF-endemic environment Gervasi and Guberti, 2021). Several studies have recommended training of pig farmers on biosecurity measures as a means of mitigating ASF. In a specifi c smallholder pig sector such as Serbian, additional costs for application of biosecurity, availability of funds are key barriers to adopt better practices Angeloni et al., 2022). In the current systems, farmers rely on cheap biosecurity and animal management measures to sustain their pig production; however, these practices are not suffi cient to stop ASF transmission (Andraud et al., 2021;Ardelean et al., 2021). Relevant stakeholders need to be educated about implementation of biosecurity measures in an eff ort to mitigate the risks (Mutua and Dione, 2021). Further, backyards could facilitate the introduction of ASFV from wild boar population to domestic pigs and vice versa (Gervasi and Guberti, 2021). Low biosecurity farms and the human factor that creates link to wild boars around the villages are deemed to be the most dangerous combinations for the spread and persistence of ASFV in domestic pig sector (Nešković et al., 2021). Indirect contact through visiting the yard by the neighbours or via shared diff erent mechanical equipment or vehicles cannot be excluded. Th us, direct or indirect contact to contaminated fomites, which entered the stable via human activities, is regarded as the most likely source of infection (Zani et al., 2019;Bellini et al., 2021). In the case of ASF outbreak in one backyard, the whole village needs to be regarded as one epidemiological unit. Th e clinical phase starts usually aft er an incubation period of about 3-5 days (Blome et al., 2020), which would be the earliest time point during the ASF infection when the owner might suspect that one of his pigs is sick. In the backyard context, detecting suspect animals depends nearly exclusively on the pig owner. However, ASF could remain unreported for longer period due to a constant supply of susceptible pigs (Liu et al., 2021). Th e slow spread of the disease from pig to pig hampers early disease detection as it leads to initially low mortalities (Schulz et al., 2019).
Outdoor keeping, semi-free range or free range pigs is common in some regions in Serbia (Prodanov- Radulović et al., 2015). Th is type of pig keeping represents one of the weakest links in the biosecurity chain and the biggest risk factor for ASF introduction. Th e interaction between wild boars and pigs can prolong ASFV circulation, as observed in Sardinia (Cappaia et al., 2018). Th e free-ranging pigs can act as a bridge in transmitting ASFV between wild boars and domestic pigs (Sauter-Louis et al., 2021; Ardelean et al., 2021). Above all, allowing domestic pigs to roam freely is a concern, for not only ASF but also for diseases of public health importance (Petrović et

CONCLUSION
Pig production in Europe is highly heterogeneous with diff erent biosecurity standards. However, in the pig systems like the one in Serbia, farmers rely on cheap biosecurity and animal management measures, which are oft en not suffi cient to prevent or control ASF. Farmers still have signifi cant knowledge gaps in view of ASF and practice various risky behaviours that might favour disease spread. Th e ongoing practices of natural mating, home-slaughtering and swill feeding can be identifi ed as main challenging biosecurity risk factors.
ASFV has a history of more than 100 years worldwide. It is anticipated that it will continue to threaten the pig industry in countries around the world for a long time in the future. Our work reviewed the main risk factors involved in the introduction and spread of ASF in Serbia and this information can be relevant in assessing the risk level of diff erent holdings in order to plan specifi c preventive measures. We can conclude that diff erent types of risks aff ect different types of farming systems, and they need to be considered when preparing a biosecurity program. In the future, the study with insights into the ASF knowledge of backyard farmers can be helpful to identify predominant risky practices carried out by them. Th ese insights may help to better understand the role of backyard farmers in the ASF epidemic in Serbia and to improve future evidence-based policies, including the development of new public awareness activities.