Clonorchiasis and opisthorchiasis: epidemiology, transmission, clinical features, morbidity, diagnosis, treatment, and control

SUMMARY

Clonorchis sinensis, Opisthorchis viverrini, and Opisthorchis felineus are important liver flukes that cause a considerable public health burden in eastern Asia, southeastern Asia, and eastern Europe, respectively. The life cycles are complex, involving humans, animal reservoirs, and two kinds of intermediate hosts. An interplay of biological, cultural, ecological, economic, and social factors drives transmission. Chronic infections are associated with liver and biliary complications, most importantly cholangiocarcinoma. With regard to diagnosis, stool microscopy is widely used in epidemiologic surveys and for individual diagnosis. Immunologic techniques are employed for screening purposes, and molecular techniques facilitate species differentiation in reference laboratories. The mainstay of control is preventive chemotherapy with praziquantel, usually combined with behavioral change through information, education and communication, and environmental control. Tribendimidine, a drug registered in the People’s Republic of China for soil-transmitted helminth infections, shows potential against both C. sinensis and O. viverrini and, hence, warrants further clinical development. Novel control approaches include fish vaccine and biological control. Considerable advances have been made using multi-omics which may trigger the development of new interventions. Pressing research needs include mapping the current distribution, disentangling the transmission, accurately estimating the disease burden, and developing new diagnostic and treatment tools, which would aid to optimize control and elimination measures.

INTRODUCTION

In the Opisthorchiidae family, the three liver flukes, Clonorchis sinensis, Opisthorchis viverrini, and Opisthorchis felineus, are of considerable public health relevance since they cause high infection rates and severe morbidity and mortality (1, 2). C. sinensis was first found in 1874 in a male Chinese in Calcutta, India, by J. F. P. McConnell (3). O. felineus was documented for the first time in 1884 by S. Rivolta in a cat and, 8 years later, by K. Winogradoff in humans in Tomsk, Russia (4). O. viverrini was first described in a civet cat by M. J. Poirier in 1886, followed by documented human infections in Chiang Mai, Thailand, by R. T. Leiper in 1915 (5). However, an infection with these liver flukes likely dates back much longer, as evidenced by eggs detected in ancient excavated corpses and the environment in eastern Asia and Russia (6–10). Indeed, C. sinensis eggs were detected in ancient corpse buried over 2,000 years ago in the People’s Republic of China (8).

The three liver fluke species have complex life cycles with definite hosts (i.e., humans and some animal reservoir hosts) and two kinds of intermediate hosts (i.e., freshwater snails and freshwater fish), characterized by an alternation of sexual and asexual reproduction in the different hosts (11–13) (Fig. 1). Hermaphroditic adult worms parasitize in bile ducts. The life span can exceed 10 years, exemplified by a documented case of C. sinensis of 26 years (14). Eggs with miracidia are discharged in human feces and are ingested by freshwater snails where they hatch. Then, sporocysts, rediae, and cercariae are produced through asexual reproduction. In C. sinensis, cercariae are released after approximately 3 months (15), while it takes 6–8 weeks in O. viverrini (16). Cercariae attach to freshwater fish, encyst in the subcutaneous tissues or muscles, and develop into mature metacercariae after approximately 6 weeks. Infective metacercariae enter human bodies when raw or undercooked freshwater fish are consumed, with excysted larvae from metacercariae entering bile ducts and developing into adult worms. Approximately 4 weeks later, eggs are produced. In humans, an adult liver fluke usually lays around 3,000–4,000 eggs per day (17, 18).

Fig 1.

(A) Life cycle of liver flukes. [Adapted from reference (13) with permission from Elsevier; the illustrations of adult C. sinensis, O. viverrini, and O. felineus were redrawn from illustrations in reference (11) with permission from Elsevier.] (B) Adult worm of C. sinensis (recovered from mice) (courtesy of Liu Liu and Hao Zhang, Qiqihar Medical University; used with permission). (C) Adult worm of O. viverrini (recovered from hamster) [reproduced from reference (20) with permission from Elsevier]. (D) Egg of C. sinensis (expelled from adult worms recovered from rat) [courtesy of Ting-Jun Zhu, National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention (Chinese Center for Tropical Diseases Research); used with permission]. (E) Cercaria of O. viverrini (shed from Bithynia siamensis siamensis) [reproduced from reference (21), which is published under a CC BY 4.0 license]. (F) Metacercaria of C. sinensis (digested from Pseudorasbora parva) (courtesy of Ting-Jun Zhu, used with permission).

Adult liver flukes show similar morphology among the three species, characterized by flattened and lance shape (11, 19) (Table 1). The differences in position of the testes and the arrangement of vitelline glands aid species differentiation (11, 20). C. sinensis has branched testes, while both Opisthorchis species have lobed testes (11, 20). Additionally, C. sinensis has continuously distributed vitelline glands, while Opisthorchis have clustered vitelline glands. The main morphologic differences between the two Opisthorchis species are the presence of deeper lobulation and a greater extremity of the testes and no transversely compressed patterns of vitelline follicles in O. viverrini. Eggs of liver flukes present oval sharp, with a convex operculum resting on the “shoulders” at the smaller end of the egg and a small knob at the opposite end (19). The mature cercaria, consisted of a body and tail, looks like a bent tobacco pipe, concaving on the ventral side and bent outward on the dorsal side (21). The metacercaria is embedded in a double-walled cyst, which usually presents oval shape (11, 20).

TABLE 1.

Sizes of human liver flukes in different stages

Stage	Clonorchis sinensis	Opisthorchis viverrini	Opisthorchis felineus
Adult (19)	10–25 mm × 3–5 mm	5–10 mm × 1–2 mm	7–12 mm × 2–3 mm
Egg (19)	27–35 µm × 11–20 µm	19–30 µm × 10–20 µm
Cercaria (11)	Body: 154 µm × 75 µm; tail: 392 µm × 26 µm
Metacercaria (11)	140 µm × 113 µm	201 µm × 167 µm	252–320 µm × 165–225 µm

In the recently published roadmap for neglected tropical diseases (2021–2030), put forth by the World Health Organization (WHO), an intensified control was proposed for hyperendemic areas of foodborne trematode infections. Three key areas are highlighted in the roadmap, namely, (i) a better scientific understanding, (ii) improved diagnostics, and (iii) effective interventions (2). To fully capture progress on these aspects, this review comprehensively summarizes the knowledge on clonorchiasis and opisthorchiasis with a focus on epidemiology, transmission, clinical features and morbidity, diagnosis, treatment, and control. Finally, we put forward a series of unresolved questions and discuss research needs in these six areas for improved management, control, and even elimination of these human diseases.

EPIDEMIOLOGY

Human liver fluke infections are endemic in Eurasia, with at least 680 million people at risk of infection according to estimates published in 2005 (22). C. sinensis is endemic in eastern Asia, O. viverrini in southeastern Asia, and O. felineus in eastern Europe (Fig. 2). In 1995, WHO provided the first estimates of the population infected with the three liver flukes: 17.34 million (7.01 million of C. sinensis infection, 8.74 million O. viverrini infection, and 1.58 million O. felineus infection) (23). It is estimated that 27.21 million people are infected with one of these species nowadays (Table 2) (23–26).

Fig 2.

Endemic status of clonorchiasis and opisthorchiasis.

TABLE 2.

Global distribution of clonorchiasis and opisthorchiasis

	Estimated population infected	Hot spots
Clonorchis sinensis
People’s Republic of China	10.82 million (24)	Southeastern region along the Pearl River covering Guangdong and Guangxi provinces, northeastern region along the Songhuajiang River covering Heilongjiang and Jilin
South Korea	1.42 million (25)	Southern part along the major rivers including Nakdong, Seomjin, Geum, and Yeongsan
Vietnam	1.00 million (23)	Northern areas, especially in Nam Dinh and Ninh Binh provinces
Russia	3,000 (23)	Part of far east area
Subtotal	13.24 million
Opisthorchis viverrini
Thailand	6.71 million (26)	Northeastern areas, especially in Buengkarn, Sakon Nakhon, Kalasin, Roi Et, and Nakhon Phanom provinces
Lao People’s Democratic Republic	2.45 million (26)	Southern areas, especially in Savannakhet, Khammouane, Champasak, and Saravane provinces
Cambodia	1.00 million (26)	Southern areas, especially in Kampong Thom and Kampong Cham provinces
Vietnam	2.07 million (26)	10 adjacent provinces in central and southern Vietnam
Myanmar	Not available	Cases detected locally in Lower Myanmar
Subtotal	12.39 million (26)a
Opisthorchis felineus
Russia	1.22 million (23)	Western Siberia in the Ob River basin, including Khanty-Mansiysk Autonomous Okrug, Yamalo-Nenets Autonomous Okrug, Tyumen Oblast, Tomsk Oblast, Omsk Oblast, and Novosibirsk Oblast
Ukraine	0.31 million (23)	Sumy and Poltava regions distributing in Dnieper River basin
Kazakhstan	0.05 million (23)	North and northeast areas drained by the Irtysh River and tributaries, especially in Pavlodar province
Byelorussia	Not available	Brest, Gomel, and Grodno provinces
Italy	Not available	Several outbreaks due to fish from two lakes Bolsena and Bracciano in central Italy
Subtotal	1.58 million (23)
Total	27.21 million

^a This number is not equal to the sum of data in Thailand, Lao People’s Democratic Republic, Cambodia, and Vietnam because Bayesian spatial-temporal joint models were applied in this study (26).

Clonorchis sinensis

C. sinensis is endemic in the People’s Republic of China, South Korea, northern Vietnam, and the far east of Russia, with an estimated 13.24 million people infected (Table 2) (23–25). The People’s Republic of China harbors the largest number of C. sinensis cases, with an estimated 10.82 million cases, due to a prevalence of 0.84% in 2015 based on data from a national survey and routine surveillance (24). The largest endemic areas in the People’s Republic of China are located in the southeastern parts along the Pearl River covering Guangdong and Guangxi provinces (27, 28). Heilongjiang and Jilin in the northeastern part of the country form another important endemic region, while high prevalence has been reported along the Songhuajiang River (29, 30). Historically, C. sinensis was also endemic in the central part of the People’s Republic of China, including the provinces of Anhui, Henan, and Shandong, where cases are only occasionally reported nowadays (31, 32).

An estimated 1.42 million people were infected with C. sinensis in South Korea in 2020, based on a prevalence of 2.77% captured from a spatial statistical analysis (25). High prevalence has been recorded in the southern parts along the major rivers, including Nakdong, Seomjin, Geum, and Yeongsan (33). C. sinensis is endemic in over 20 provinces in northern Vietnam, and high prevalence (>20%) has been reported in the provinces of Nam Dinh and Ninh Binh (34, 35). Approximately 1 million people are affected by C. sinensis there, but a clear picture on the epidemiology remains elusive, as national surveys have yet to be conducted (23). In addition, several thousand cases with C. sinensis are likely to occur in the far eastern part of Russia (36).

Historically, C. sinensis was endemic in Japan, with high prevalence observed in Akita, Chiba, Ibaraki, Okayama, Shiga, and Tokushima, while only few cases are documented nowadays (37). C. sinensis was also once reported from central Thailand (38). However, it is not clear whether C. sinensis indeed occurs in Thailand (39, 40).

Opisthorchis viverrini

O. viverrini is endemic in southeastern Asia. Bayesian spatial-temporal models were recently applied to map the distribution of O. viverrini in southeastern Asia, which revealed that an estimated 12.39 million people were infected in 2018, more than half in Thailand (6.71 million), followed by Lao People’s Democratic Republic (2.45 million), Vietnam (2.07 million), and Cambodia (1.00 million) (Table 2) (26).

Repeated cross-sectional surveys demonstrated a decline of O. viverrini infection in Thailand, with an updated crude prevalence of 2.2% (356 infections among 16,187 people surveyed) in 2019. The highest prevalence was observed in the northeastern part of the country (5.0%), followed by areas in the northern part (1.8%), while a lower prevalence was observed in the central (0.9%) and southern parts of the country (0.1%) (41). Buengkarn, Sakon Nakhon, Kalasin, Roi Et, and Nakhon Phanom provinces are hot spots of opisthorchiasis. In Lao People’s Democratic Republic, O. viverrini infections primarily occur in the southern parts of the country, specifically in the provinces of Champasak, Khammouane, Saravane, and Savannakhet (42, 43). Additionally, in the northwestern parts around the capital Vientiane, O. viverrini is also endemic (43). In Cambodia, although O. viverrini infections have been documented in nearly all provinces with the exception of Kampong Speu, Kep, and Prey Veng, major endemic areas were documented in the central parts from Preah Vihear province in the North to Takeo province in the South (44, 45). Kampong Thom and Kampong Cham provinces are hot spots (44, 45). O. viverrini infections occur in 10 adjacent provinces in the central and southern parts of Vietnam, extending from Thua Thien Hue province in the North to Dac Nong province in the South (34). Recently, O. viverrini infections were reported in Myanmar. In 2015 and 2016, O. viverrini was detected in rural populations in three regions, namely, Bago region, Mon state, and Yangon region of Lower Myanmar (46). Meanwhile, a large survey in late 2015 revealed a low endemicity of O. viverrini infection in Yangon (0.7%, 14 infections among 2,057 surveyed individuals) (47).

Opisthorchis felineus

O. felineus is endemic in Russia, Kazakhstan, and Ukraine, while foci have also been documented in Belarus and several outbreaks were investigated in Italy (Table 2) (23, 48). However, no systematic surveys have been implemented in endemic countries.

An estimated 1.58 million people were infected in the 1990s based on expert opinion, while accurate and updated data are still lacking (23). Russia is the predominant endemic area for O. felineus, where approximately 1.22 million people were infected in 1990. High endemic areas are concentrated in western Siberia in the Ob River basin, including Khanty-Mansiysk Autonomous Okrug, Yamalo-Nenets Autonomous Okrug, Tyumen oblast, Tomsk oblast, Omsk oblast, and Novosibirsk oblast (49, 50). Between 2011 and 2013, an incidence of 25 cases per 100,000 people was documented based on surveillance data (49). In the Ukraine, O. felineus infection is endemic in Sumy and Poltava regions in the Dnieper River basin (51). Approximately 312,000 people were infected in 1990 (23). In Kazakhstan, O. felineus infection is mainly endemic in the northern and northeastern parts that are drained by the Irtysh River and its tributaries (52). Pavlodar province is the setting with the highest prevalence. A focus of O. felineus infection in the northwest in the Ural River basin is also documented. Approximately 49,000 people were estimated to be infected in 1990 in Kazakhstan (23), while 1,000–3,000 cases were reported annually from 1997 to 2011 (52). In Brest, Gomel, and Grodno provinces of Belarus, opisthorchiasis was once reported (53). However, no updated information is available. High prevalence of O. felineus is reported in fish from two lakes, Bolsena and Bracciano in central Italy. Indeed, consumption of fish from these two lakes caused several O. felineus outbreaks in Italy (48, 54). Human liver fluke infections are occasionally reported in other parts of the world because of immigrants from endemic countries, consumption of raw or undercooked freshwater fish in endemic areas when traveling, or importation from endemic areas (55–57).

DETERMINANTS OF TRANSMISSION

Human behavior

Many factors influence the transmission of liver fluke infections, of which human behavior, specifically the habit of consumption of raw aquatic food, is critical (58). Differences in raw-eating behavior govern the heterogeneity of liver fluke infections in the population, both spatially and temporally (59). Usually, children ingest raw or undercooked freshwater fish infrequently, while the proportion and frequency increase with age, particularly in males (60). Thus, high infection and intensity of liver fluke are common in elderly males (33, 41, 61). However, the infection reaches a plateau in middle-aged people and usually decreases after the age of around 60 years. Raw-eating habits result from a complex interplay between social, economic, and cultural determinants. Raw-eating behavior is influenced by cultural awareness, community identification, familial environment, and individual choice (60, 62, 63). In many parts of the People’s Republic of China, South Korea, and Vietnam, raw fish are consumed, accompanied with ginger, garlic, and other side dishes (64–66). In southeastern Asia, raw fish spicy salad and fermented fish are often consumed (62, 67). In Russia, smoked, sun-dried, frozen fish and stock fish are ingested (68). Raw-eating behavior is passed on from one generation to the next within the family and community (60, 67). Sharing and consumption of raw fish at social events promote the transmission at community level (69, 70), while interaction among family members might trigger the establishment of raw-eating behavior in those who had not done so before (60, 71). Children, especially boys, are prone to take on the behavior from their parents (60). Alcohol is usually consumed while ingesting raw fish, which also contributes to specific age and gender patterns (66, 72). While alcohol might shelter fishery smell, many people hold the wrong belief that it kills metacercariae in fish. Thus, raw-eating behavior is mingled with misconception and lack of awareness (63).

People’s economic status has been associated with the prevalence of liver fluke infection. Economic development promotes the development of aquaculture and provision of freshwater fish. In some endemic areas, raw fish are often consumed at restaurants (65). This might explain the high endemicity of clonorchiasis in the Pearl River area of Guangdong, an important economic center in the People’s Republic of China (27). In Guangxi, a neighboring province of Guangdong, a significant increase in the prevalence of clonorchiasis has been documented over the past decades because more people attempt to ingest raw freshwater fish (28). However, in other areas, lower economic status is associated with liver fluke infection (68). High prevalence is often found in people with a low educational attainment, such as fishermen and farmers (35, 42) but is also reported in highly educated people, including medical workers and businessmen (27, 28).

High re-infection rates cause persistence of human liver fluke infections. Re-infection is governed by the habit of ingesting raw or undercooked freshwater fish, which depends on disease endemicity, and the demography and infection intensity of individuals (73–75). In general, the higher the endemicity, the more rapid re-infection occurs (73, 75). In highly endemic areas, the prevalence could even bounce to original levels within 1 or 2 years after ceasing treatment and other control measures (73). Males and elderly people show a higher tendency to re-infection (73–75). Those with heavy infection intensity before treatment are particularly prone to re-infection, and they are also more likely to show heavy infection intensity during re-infection (73, 75). High re-infection sometimes leads to the occurrence of an interesting phenomenon, namely, higher re-infection rates of those with previous treatment history, because particularly individuals with frequent treatment are prone to continue ingesting raw or undercooked freshwater fish (76).

Freshwater snails

The distribution of specific freshwater snails determines the range of liver fluke infections. Thus far, at least 13 different species of freshwater snails have been recognized as first intermediate hosts of C. sinensis, most of which belong to the Bithyniidae family (77, 78). However, only three species (i.e., Parafossarulus striatulus, Alocinma longicornis, and Bithynia fuchsiana) were confirmed by experimental infection and/or molecular evidence, while the other species were defined as intermediate host just through morphologic examination (78). Since a large number of diverse foodborne trematodes can infect freshwater snails, an accurate identification by morphology is challenging (79, 80). For example, it has recently been confirmed that Melanoides tuberculata, previously thought to be an important intermediate host snail in the transmission of C. sinensis in Vietnam, is not a suitable host of C. sinensis based on observational, molecular, and experimental studies (78). Three Bithynia taxa (i.e., Bithynia funiculata, B. siamensis goniomphalos, and B. s. siamensis) act as first intermediate hosts of O. viverrini in Southeast Asia (81). B. leachi, B. troscheli, B. inflata, and B. tentaculata are recognized as first intermediate hosts of O. felineus (82).

Snails belonging to the Bithyniidae family show preferences for habitats characterized by slow-flowing muddy rivers, lakes, ponds, canals, and swamps (83). However, different snail species have their own preferences, although they sometimes are found in the same habitat. P. striatulus snails are found in ponds, ditches, streams, and swampy areas, characterized by abundant water grasses and organic materials (84). B. funiculata snails prefer habitats with clear and shallow water (depth: <30 cm) (85). B. s. goniomphalos snails prefer slightly saline water and nitrite-nitrogen, while large snail populations have been observed in rice fields, ponds, road-side ditches, canals, and lakes during the rainy season (86, 87). In the cold and dry season, snails can bury themselves in the mud and survive for prolonged periods. In the northeastern part of the People’s Republic of China, about one-third of P. striatulus snails survives after hibernation during the winter months (November–February) (84). A study demonstrated that in the northeastern part of Thailand, approximately half of B. s. goniomphalos snails survived the entire cool-dry season (November–February) with increasing mortality rates from the beginning to the end of the cold season (88). The mortality of snails is enhanced if droughts persist, with an over 90% mortality reported after a 17-month drought in northeastern Thailand (89).

The prevalence of infection in the first intermediate hosts is usually very low (<1%) (21). However, high prevalence has occasionally been reported, e.g., O. viverrini in B. s. goniomphalos in some areas in Thailand (6.9%, 21/303) and Lao People’s Democratic Republic (8.4%, 17/203) and C. sinensis in P. striatulus in some areas in the People’s Republic of China (12.3%, 25/204), as determined by cercarial shedding (90, 91). Snail infection is influenced by environmental factors (e.g., temperature, precipitation, and hours of sunshine) and human activities (92). Hence, there is often a pronounced seasonality of prevalence, which varies by species and/or locations. High prevalence of O. viverrini in snails was shown in the cool-dry season (November–February) after the rainy season (June–October) (92–94). However, the infection of C. sinensis in P. striatulus was observed to be high in the summer months (June and July) in the northeastern part of the People’s Republic of China (95). The role of temperature in snail infections is complex. On the one hand, increasing temperature causes high mortality of miracidia and low success in egg hatching (96). On the other hand, it was also demonstrated that a 1°C increase (in the temperature range from 16°C to 37°C) was associated with a 5.4% increase in the odds of infection with O. viverrini in B. s. goniomphalos snails, peaking at 34°C (97). There are reports that female snails have a higher prevalence of infection than males, as shown for P. striatulus snails infected with C. sinensis and B. s. goniomphalos snails infected with O. viverrini (88, 95). Infection also depends on the size of snails, although controversial findings are reported in the literature. In Khon Kaen province, Thailand, B. s. goniomphalos snails with a large size (>8 mm in length) had a higher prevalence with O. viverrini compared to their smaller counterparts (88). However, in Vientiane province in Lao People’s Democratic Republic, during the rainy and cool-dry seasons, there was a higher prevalence in small B. s. goniomphalos snails (<8 mm in length) compared to larger specimens, but there were no significant differences during the hot-dry season (March–May) (93). In Sakon Nakhon province, Thailand, higher prevalence was shown in large size B. s. goniomphalos in the hot-dry season, while the opposite was observed in the rainy and cool-dry seasons with higher prevalence in small size snails (98). Experiments pursued with O. viverrini eggs in B. funiculata and B. s. siamensis demonstrated that young snails (aged 1–3 months) bred in the laboratory appear to be more susceptible than field-collected mature ones, while the prevalence in B. s. goniomphalos is low in both young and mature snails (99).

High numbers of O. viverrini cercariae are released from B. s. goniomphalos in the hot-dry season (93, 94, 98). While cercariae do not emerge during darkness, low-intensity light could induce a release (100, 101). The peak of cercarial shedding usually occurs between 08:00 and 14:00 h, which differs among seasons and varies by areas. For example, in Vientiane province, cercarial shedding from B. s. goniomphalos peaks between 08:00 and 10:00 h during the hot-dry season and between 12:00 and 14.00 h during the rainy and cool-dry seasons (93). In Sakon Nakhon province, Thailand, the peak of cercarial emergence occurs between 08:00 and 10:00 h during the rainy and cool-dry seasons and between 10:00 and 12:00 h in the hot-dry season (98). Shell size of snails is usually positively correlated with the cercarial output per day (92, 94).

Freshwater fish

Thus far, approximately 100 species of freshwater fish have been recognized in the transmission of C. sinensis, more than 50 species in the case of O. viverrini, and some 20 species in O. felineus (102–104). Of note, the fish predominantly belong to the family Cyprinidae (carp). Additionally, several shrimp species have been reported as host for C. sinensis (105, 106).

The infection in freshwater fish presents variation, including the species and size of fish, sources of areas, types of water, and seasons, which is determined by the interaction among environmental factors, liver fluke prevalence in definite hosts, distribution of first intermediate hosts, and biology and behavior of freshwater fish (104). Although many fish species could harbor C. sinensis and Opisthorchis spp., the infection in terms of prevalence and intensity varies significantly. Small size fish are usually highly infected, such as Pseudorasbora parva and Pungtungia herzi. Hence, it has been suggested that these fish could serve as index species to assess the risk of transmission (106–108). In the laboratory, it was found that Barbonymus gonionotus is more susceptible to be infected with O. viverrini as a function of growing age and size (109). In a study done with natural bred fish, higher prevalence with O. viverrini was detected in larger sized fish of Cyclocheilichthys armatus and Puntius brevis, while no association was observed in Hampala dispar (110). Another study found that small Cyprinus carpio fish were more susceptible to be infected with other type of trematodes (Heterophyidae) (111). The variation of infection in different species of freshwater fish is probably attributed to a complex interplay of biological susceptibility to infection due to physiology, immunology, and nutrition, coupled with habitat characteristics (112, 113). Fish treated with immunosuppressant (e.g., prednisolone) showed high susceptibility to infection and likelihood of development of infective metacercariae with O. viverrini (112). The activity of superoxide dismutase enzyme involved in defense of parasite infection is reduced in fish treated with an immunosuppressant (112). In terms of habitat characteristics, the species of fish acting as intermediate hosts of liver flukes generally have a broad activity space (113). Usually, high prevalence with C. sinensis is reported in summer months, while with O. viverrini in late rainy and cool-dry seasons (104, 110, 114). Little is known on the morbidity of liver fluke infections on fish, except one laboratory study, which demonstrated that B. gonionotus infected with O. viverrini cercariae from Nam Ngum River in Lao People’s Democratic Republic suffered from higher mortality compared to uninfected fish (115). However, the same species infected with O. viverrini cercariae from the Songkram River in Thailand showed no significant change in mortality compared to the control.

Animal reservoirs

Piscivorous animals act as reservoir hosts in liver flukes. Cats and dogs are the most important species in terms of prevalence and abundance (116–120). The prevalence is usually higher in cats compared to dogs. Infection level is related to the source, feed foods, and age of animals (117, 118). Higher prevalence is usually observed in older animals (118). The prevalence in cats and dogs is usually higher in highly endemic areas with liver fluke infection in humans, but high prevalence in animals is also reported in areas with low infection and even without infection in humans, which indicates the existence of animal circulation (117). Movement and human-induced transfer of animals could spread an infection into other areas (121). In O. felineus, up to 15 wild carnivorous mammals might also serve as hosts (82). Pigs infected with C. sinensis and O. viverrini have also been observed (116, 119). Ducks had been suggested as reservoir of O. viverrini, but recent molecular phylogenetic evidence refuted this claim (122–124).

CLINICAL FEATURES AND MORBIDITY

Clinical features

Infection with human liver flukes causes an array of symptoms and morbidities, in which the hepatobiliary system is predominantly involved. Symptoms and morbidity are related to the infection intensity and duration of infection (125). Early infection leads to nonspecific symptoms, including asthenia, lassitude, nausea, indigestion, anorexia, diarrhea, headache, and abdominal discomfort, especially in the right upper quadrant (54, 126, 127). Jaundice, icteric conjunctivae, hepatomegaly, and tenderness on liver are important physical signs (54, 126, 127).

Morbidity

Chronic infection causes organ damage, including thickening and irregularities of the gallbladder wall, presence of sludge and enlargement of the gallbladder, dilatation and periductal fibrosis of bile duct (68, 72, 128–130). Subsequent complications include cholangitis, cholecystitis, and cholelithiasis (68, 72, 131, 132). Cholelithiasis is frequent, and both stones in gallbladder and bile duct have been associated with an infection (72, 133). Pigment stones are the major stone type caused by liver flukes (134, 135) (Fig. 3). Fatty liver is also a common symptom in those with liver fluke infections (129, 130). This observation has been confirmed by a cross-sectional study in a C. sinensis endemic area (72). Human liver fluke infections have also been associated with renal pathology (136–139). Estimated glomerular filtration rate declines in C. sinensis and O. viverrini infection. Both C. sinensis and O. viverrini infections demonstrate a negative impact on the development of children, including stunting, underweight, and low body mass index, and severe infection might cause dwarfism (140–142).

Fig 3.

Morbidities associated with liver fluke infections. (A) Clonorchis sinensis eggs (white arrow) in gallbladder stone under scanning electron microscopy [reproduced from reference (135)]. (B) Liver with cholangiocarcinoma (white arrow) and Clonorchis sinensis (black arrow) [reproduced from reference (12) with permission from Elsevier].

The most severe and final sequela of liver fluke infection is cholangiocarcinoma. In 1994, O. viverrini was classified as carcinogen (Group 1) to humans, due to a causative role in cholangiocarcinoma, while C. sinensis was classified as a probable carcinogen (Group 2A) to humans (143, 144). In 2009, C. sinensis was re-adjusted and also classified as definite carcinogen (Group 1) to humans (145, 146). In a case-control study, C. sinensis infection based on radiologic evidence was significantly associated with all three subtypes of cholangiocarcinoma (intrahepatic, hilar, and distal extrahepatic) (147). A significant relationship was also found between intensity of infection and cholangiocarcinoma in O. viverrini (148). Odds ratios varied in different studies: it was as high as 13.6 in C. sinensis based on stool examination (149) and 27.1 in O. viverrini infection based on antibody level (150). Evidence provided through meta-analyses reported an odds ratio between 4.5 [95% confidence interval (CI), 2.6–7.7] and 6.1 (95% CI, 4.4–8.6) in C. sinensis (61, 151) and 4.4 (95% CI, 1.0–18.6) in O. viverrini (151), while the combined value was 4.8 (95% CI, 2.8–8.4) (152). The yearly incidence of cholangiocarcinoma attributed to C. sinensis infection is estimated to be 25 per 100,000 in females and 35 per 100,000 in males (61).

Potential related morbidity

O. felineus is not classified as carcinogen, but there also exists potential association from ecologic studies and case reports (49, 153). A recent case-control study adds further evidence in the potential carcinogenicity of O. felineus. Among 39 cholangiocarcinoma cases, the proportion with O. felineus eggs in feces was 61.5%, while it was 43.1% in 160 controls, resulting in an odds ratio of 2.5 (95% CI, 1.2–5.5) (154). C. sinensis infection has also been related to primary hepatocellular carcinoma. Besides some case reports, a case-control study provided evidence in which a C. sinensis infection was 16.4% (73/444) in those with primary hepatocellular carcinoma, while it was 2.4% (12/500) in the control group, yielding an odds ratio of 8.0 (95% CI, 4.3–14.9) (155). Additionally, case reports also documented a potential association of human liver fluke infections with other morbidities, such as pancreatitis (156), eosinophilic pneumonia (157), and carcinoma of the gallbladder (158) in C. sinensis infection and heart diseases in O. felineus infection (159).

Several studies found an association between human liver fluke infections with allergy (160–164). Indeed, C. sinensis infection is associated with a higher responsiveness to common aeroallergens in skin prick test but does not show an association with allergic diseases (e.g., wheezing, airway hyperresponsiveness, asthma, or allergic rhinitis) (164). However, a higher risk of rhinitis is demonstrated in individuals with O. viverrini infection, compared to those without an infection (162). A study considering both rural and urban areas showed an inconsistent correlation between O. felineus infection and allergy (161). In individuals from rural areas, the prevalence of allergic diseases (e.g., bronchial asthma, atopic dermatitis, and allergic rhinitis) was higher in individuals infected with O. felineus. In urban areas, the presence of antibodies to O. felineus was negatively correlated with atopic sensitization to common allergens in skin prick test.

Co-factors

Viral hepatitis is highly prevalent in areas endemic for human liver flukes (165). A study carried out in the southeastern part of the People’s Republic of China showed weaker liver function and higher HBV DNA titers in individuals with C. sinensis and HBV coinfection (166). Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were higher in coinfected individuals, compared to both C. sinensis mono-infection and HBV mono-infection. HBV DNA titers were also higher in coinfected individuals compared to HBV mono-infection. However, another study in the northeastern part of the People’s Republic of China demonstrated that coinfection with C. sinensis and HBV or HCV did not aggravate the manifestations of viral hepatitis B and viral hepatitis C (30). ALT and AST were similar in HBV and C. sinensis coinfected individuals compared to both C. sinensis mono-infection and HBV mono-infection. However, they were higher in individuals coinfected with HCV and C. sinensis, compared to C. sinensis mono-infection, but not in HCV mono-infection. It has been documented that O. viverrini infection is related to a higher frequency of cagA-positive Helicobacter pylori, which contributes to the development of periductal fibrosis in opisthorchiasis (167, 168). Thus, it has been argued that H. pylori may contribute to the development of cholangiocarcinoma caused by O. viverrini (169). Higher liver pathology is documented in individuals with O. viverrini and Schistosoma mekongi coinfection, specifically a higher risk in liver fibrosis and enlargement of left liver lobe (170). Some nonbiologic factors could also exacerbate the morbidity caused by liver fluke infections (171–173). Alcohol is usually consumed when ingestion of raw fish, while alcohol is also related to cholangiocarcinoma (171, 172). Additionally, fermented raw fish containing N-nitroso compounds, e.g., N-nitrosodimethylamine, may also promote the damage and even cholangiocarcinoma (173).

Liver flukes and liver transplantation

Over 20 cases of donor liver containing liver flukes upon transplantation have been documented, of which most donor livers contained C. sinensis (174, 175). Most of the cases were reported from a single hospital in Tianjin, the People’s Republic of China (176). Out of 3,288 donor livers from May 2003 to December 2009, 14 were detected with C. sinensis (176). At present, definitive criteria for the management of donor livers characterized by a liver fluke infection do not exist, and these livers qualify for transplantation. However, proper screening of donors from endemic areas with liver fluke infection is recommended (177). In case an infection is detected in living donor preoperatively, praziquantel is recommended to be administered to the donor before transplantation (178, 179). In most cases, adult flukes or eggs were detected intraoperatively (174–176, 179). Although low-temperature perfusion and cold preservation for the donor livers are considered to kill the worms, praziquantel is still used for the recipients postoperatively (174–176, 179). Regular follow-up of the recipients is important to monitor the occurrence of complications (180). Additionally, life-long surveillance for the development of malignancy is needed because both C. sinensis and O. viverrini could cause cholangiocarcinoma, while prolonged immunosuppression in transplanted receipts could aggravate certain malignancies (177, 180).

DIAGNOSIS

The diagnosis of liver fluke infections not only involves humans but also intermediate hosts and animal hosts (12, 181). In humans, different diagnostic methods are available, including parasitologic methods, immunologic techniques, molecular methods, and imaging. These methods will be summarized below.

Traditional parasitologic methods

Detection of liver fluke eggs in feces is the most widely used diagnostic approach and usually considered as the “gold standard.” Direct fecal smear, the Kato-Katz thick smear method, and the formalin ethyl-acetate (or ether) concentration technique (FECT) are often applied (182, 183). The Kato-Katz thick smear method and FECT are most popular and widely used in epidemiologic surveys and surveillance because they show high sensitivity and allow for a semi-quantitative appraisal of infection intensity (32, 33, 41, 182, 183). The Kato-Katz thick smear method is relatively simple and cheap, while chemical agents and specific laboratory equipment are needed in FECT (e.g., centrifuge). An advantage with FECT is that feces can be stored for a longer time, and intestinal protozoa can concurrently be detected by FECT. No clear conclusion can be drawn related to which method shows superior sensitivity. Some studies reported similar sensitivity of the two methods (184), while other studies showed higher sensitivity of the Kato-Katz thick smear method (185), or FECT (183). However, the sensitivity of fecal examination techniques is highly influenced by the infection intensity, with lower sensitivity in low-intensity infection (182, 186). To overcome this limitation, multiple thick smears, repeated fecal samples, and a combination of several fecal examination techniques have been suggested (184–186). The quality of fecal examination is highly dependent on the availability of experienced technicians. Recently, an automatic vision-based examination system has also been developed, which, however, showed low sensitivity and, hence, needs further improvement to decrease false-negative results (187).

During surgery and clinical examination, the detection of eggs in bile is a sensitive diagnostic marker (188). Detection of eggs in gallstone is also useful (189). The morphology of eggs of the three species of liver flukes covered in this review is similar, and they are also similar to those of other minute intestinal flukes, which usually co-occur in liver fluke endemic areas (190, 191). Hence, morphologic examination of eggs fails to differentiate between the three species, and it is challenging to differentiate them from other minute intestinal flukes. The analysis of adult worms during clinical surgery could help species identification (192, 193). Additionally, collection of worms after treatment with anthelmintic drugs is useful for definite diagnosis (194).

Immunologic techniques

Immunologic techniques are supplementary diagnostic methods for human liver fluke infections, consisting of screening in epidemiologic surveys and clinical settings. Several immunologic diagnostics have been studied and developed, of which enzyme-linked immunosorbent assay (ELISA) has become the mainstay (Table 3).

TABLE 3.

Sensitivity and specificity of different immunologic diagnostics for human liver fluke infections^r

	Targeted product	Used product	Species	Gold standard	Sensitivity	Specificitya
Serum antibody
ELISA (195)	Antibody (IgG)	CAW	C. sinensis	Stool examination	88.2% (449/509)	100.0% (163/163)
ELISA (196)	Antibody (IgG)	CAW	C. sinensis	Kato-Katz method + water washing precipitation method	81.5% (194/238)b	92.4% (157/170)c
ELISA (197)	Antibody (IgG)	CAW	C. sinensis	Stool examination and/or serologic ELISA	C. sinensis: 91.6% (76/83); O. viverrini: 97.1% (34/35)d	95.0% (38/40)
ICT (198)	Antibody (IgG)	CAW	O. viverrini	Unclear	86.6% (97/112)	96.7% (29/30)
ICT (198)	Antibody (IgG4)	CAW	O. viverrini	Unclear	75.0% (84/112)	100.0% (30/30)
ELISA (195)	Antibody (IgG)	ESA	C. sinensis	Stool examination	92.5% (471/509)	100.0% (163/163)
ELISA (199)	Antibody (IgG)	ESA	C. sinensis	Diverse methodse	80.4% (41/51)	78.6% (103/131)f
ELISA (200)	Antibody (IgG)	ESA	C. sinensis	Kato-Katz method for cases	88.6% (31/35)	97.2% (35/36)
ELISA (201)	Antibody (IgG)	ESA	O. viverrini	Stool examination	O. viverrini: 100.0% (92/92); C. sinensis: 80.0% (24/30)g	100.0% (45/45)
ICT (201)	Antibody (IgG)	ESA	O. viverrini	Stool examination	O. viverrini: 94.6% (87/92); C. sinensis: 90.0% (27/30)g	100.0% (45/45)
ELISA (202)	Antibody (IgG)	ESA	O. felineus	Stool examination	100.0% (144/144)	95.5% (105/110)
ELISA (203)	Antibody (IgG4)	Antigen (cathepsin L protease)	C. sinensis	Stool examination for cases	62.5% (15/24)	95.8% (23/24)
ELISA (200)	Antibody (IgG)	Antigen (Cs1)h	C. sinensis	Kato-Katz method for cases	94.3% (33/35)	94.4% (34/36)
ELISA (204)	Antibody (IgG)	Antigen (CsGSTo1)i	C. sinensis	Stool examination for cases	C. sinensis: 90.0% (108/120); O. viverrini: 95.4% (83/87)d	100.0% (40/40)
ELISA (204)	Antibody (IgG)	Antigen (CsGSTo2)j	C. sinensis	Stool examination for cases	C. sinensis: 89.2% (107/120); O. viverrini: 98.9% (86/87)d	100.0% (40/40)
ELISA (197)	Antibody (IgG)	Antigen (CsAg17)	C. sinensis	Stool examination and/or serologic ELISA	77.1% (64/83)	100.0% (40/40)
ELISA (205)	Antibody (IgG)	Multiple antigens (Cs26GST + Cs28GST)k	C. sinensis	Stool examination for cases	76.7% (23/30)	100.0% (17/17)
ELISA (205)	Antibody (IgG)	Multiple antigens (Cs26GST + Cs28GST + CsVpB1)l	C. sinensis	Stool examination for cases	86.7% (26/30)	100.0% (17/17)
ELISA (197)	Antibody (IgG)	Fusion antigens (Cs28GST-CsAg17)	C. sinensis	Stool examination and/or serologic ELISA	76.1% (35/46)	100.0% (30/30)
Saliva antibody
ELISA (206)	Antibody (IgG)	CAW	O. viverrini	FECT	90.0% (27/30)	65.2% (15/23)
ELISA (206)	Antibody (IgA)	CAW	O. viverrini	FECT	96.7% (29/30)	4.3% (1/23)
Serum antigen
ELISA (207)	Antigen (cysteine proteinase)	Anti-recombinant cysteine proteinase IgY	C. sinensis	Stool examination for cases	85.7% (36/42)m	100.0% (20/20)
Feces antigen
ELISA (208)	Antigen (cathepsin F)	Anti-recombinant cathepsin F IgY	O. viverrini	FECT	93.3% (28/30)	73.3% (22/30)
ELISA (209)	Antigenn	Monoclonal antibodyo	O. viverrini	FECT	91.5% (387/423)p	68.5% (425/620)
Urine antigen
ELISA (210)	Antigenn	Monoclonal antibodyo	O. viverrini	FECT	89.6% (112/125)	55.6% (35/63)
ELISA (209)	Antigenn	Monoclonal antibodyo	O. viverrini	FECT	88.7% (375/423)q	69.2% (429/620)

^a Only negative was included, while the cross reactivity to other parasites was not listed.

^b 63.0% (51/81) in eggs per gram of feces (EPG) of 0–999, 92.6% (75/81) in EPG of 1,000–9,999, and 89.5% (68/76) in EPG over 10,000.

^c 90.0% (108/120) in negative individuals from same endemic area and 98.0% (49/50) in negative individuals from non-endemic area.

^d As a universal antigen, the cross reactivity to O. viverrini was demonstrated in sensitivity.

^e Stool examination, biopsy, endoscopic retrograde cholangiopancreatography, ultrasonography, or computed tomography.

^f 93.9% (92/98) in low-risk control with no history of ingestion of raw freshwater fish and 33.3% (11/33) in high-risk control with frequent ingestion of raw freshwater fish.

^g As a universal antigen, the cross reactivity to C. sinensis was demonstrated in sensitivity.

^h Cs1 contains tandem repeats in excretory-secretory antigens and crude antigens of adult worm.

ⁱ CsGSTo1, one isoform of omega-class glutathione S-transferases.

^j CsGSTo2, one isoform of omega-class glutathione S-transferases.

^k Cs26GST, 26-kDa glutathione S-transferases in C. sinensis, Cs28GST, 28-kDa glutathione S-transferases in C. sinensis.

^l CsVpB1 is 838 bp long, encoding a polypeptide of 245 amino acids that is homologous with vitelline precursor protein of invertebrates.

^m 75.0% (9 of 12) in EPG of below 1,000; 86.7% (13 of 15) in EPG of 1,000–4,999; 93.3% (14 of 15) in EPG of 5,000–9,999.

ⁿ Molecules present in the excretory-secretory antigens.

^o Monoclonal antibody-based (clone KKU505) reactive to excretory-secretory antigens of O. viverrini adult worms.

^p 89.6% (189/211) in EPG of 1–50; 83.3% (60/72) in EPG of 51–100; 98.6% (138/140) in EPG of over 100.

^q 83.4% (176/211) in EPG of 1–50; 87.5% (63/72) in EPG of 51–100; 97.1% (136/140) in EPG of over 100.

^r CAW, crude antigens of adult worm; ESA, excretory-secretory antigens; ICT, immunochromatographic test; FECT, formalin ethyl-acetate (or ether) concentration technique.

Detection of antibody in serum is the most common type, and crude antigens of adult worms, excretory-secretory antigens, and recombinant proteins have been tested. The sensitivity is high using crude antigen, while specificity is usually low, especially in settings where there is cross-reactivity between different species of liver flukes (e.g., C. sinensis and O. viverrini) and to Paragonimus spp. (195–198). The application of excretory-secretory antigens shows improved performance in terms of sensitivity, specificity, and cross-reactivity (195, 199–202). In view of the high cost and challenges in production of excretory-secretory antigens, recombinant proteins have been developed, such as cathepsin L protease (203), Cs1 (200), glutathione S-transferases (204), or CsAg17 (197). Recombinant proteins usually belong to excretory-secretory antigens and tegument because of their exposure and potential high antigenicity. To increase the sensitivity, two or more proteins are mixed (cocktail) or fused, e.g., the mix of 26-kDa and 28-kDa glutathione S-transferases of C. sinensis (205) and the fusion of 28-kDa glutathione S-transferases and CsAg17 in C. sinensis (197). Compared to the traditional ELISA, the developed immunochromatographic test kits show advantages in rapid diagnosis without a special reading device (196, 198, 201). The sensitivity of those serological tests is usually associated with infection intensity, with higher sensitivity in higher intensity infections (196). The specificity is lower in those individuals from endemic areas compared to people from non-endemic areas (196). Furthermore, the specificity is also lower in those at high risk, e.g., individuals ingesting raw freshwater fish (199). On one hand, cross-reactivity is usually high between C. sinensis and O. viverrini, which hampers species differentiation (195, 198, 205). On the other hand, a particularly high cross-reactivity indicates the role of universal antigens in the diagnosis of both species, e.g., crude antigens of adult worms (197), excretory-secretory antigens (201), and omega-class glutathione S-transferases (204). Saliva samples have also been used for diagnosis because of their ease of collection and non-invasiveness. However, the diagnostic performance in saliva is poor due to the low specificity (206). The major challenge of antibody detection is its inability to differentiate between active and past infections because of the persistence of antibody in serum or other samples after cure, which is demonstrated by the low specificity of serological tests in those individuals from endemic areas and ingesting raw freshwater fish (196, 199).

To overcome this shortcoming, techniques detecting antigens in serum are also explored, such as the detection of cysteine proteinase in serum (207). Coproantigen detection techniques have also been explored in O. viverrini, e.g., a monoclonal antibody (named clone KKU505) targeting molecules present in the excretory-secretory antigens and the chicken IgY-based coproantigen capture ELISA targeting cathepsin F (208, 211). Particularly, the former technique was applied in the field, demonstrating a high sensitivity and moderate specificity (209). The sensitivity is also higher in heavy intensity infections. KKU505 has also been used in capturing antigens in urine with a high sensitivity and moderate specificity (209, 210). Low daily variation is shown in terms of the sensitivity owing to the stability of antigens in urine (212, 213). It is worth highlighting that the performance in terms of sensitivity and specificity is also high in settings that undergo chemotherapy (214). To improve the applicability of antigen-based tests, efforts are made to decrease the time needed to run this assay and increase the limit of detection (215, 216). A portable fluorometer equipped with a smartphone camera has been developed, which promotes its potential application in the field (217).

Molecular methods

A host of molecular techniques targeting copro-DNA have been developed for diagnosis of liver fluke infections in humans. Internal transcribed spacers (ITS-1 and ITS-2) (218–221) and ribosomal RNA gene (222) of nuclear ribosomal DNA, and cytochrome oxidase subunit 1 (cox1) (221, 223–225), NADH dehydrogenase subunits (223, 226), and cytochrome b (221) of mitochondrial DNA are usually applied as target genes, while some other genes, including the long-terminal repeat of C. sinensis retrotransposon1 (CsRn1) gene (227), and pOV-A6 (228–230) are also used. Conventional PCR is mostly employed, which is mainly used for detection rather than differentiation between different liver fluke species. Some primers could be used quite universally, and similar length of base pairs (bp) is presented by all three species of liver flukes (218). In such cases, sequencing is needed to differentiate the species. In most studies, the molecular method was developed to target one particular species, but the differentiation capacity has yet to be evaluated and reported. Multiplex PCR shows the capacity to differentiate between three species of liver flukes as well as other minute flukes (e.g., Haplorchis taichui and Metorchis bilis) (225, 231, 232). PCR-linked restriction fragment length polymorphism can also differentiate among species (38, 40). A set of primers targeting ITS2 demonstrates 375 bp in O. viverrini, 381 bp in C. sinensis, and 526 bp in H. taichui, when AcuI digests C. sinensis into 286 bp and 95 bp (38), and FauI digests O. viverrini into two bands of 129 bp and 247 bp (40). In real-time PCR, the cycle-threshold values reflect parasite-specific DNA loads (233, 234). Loop-mediated isothermal amplification has also been developed, which is quite simple and produces rapid results (224, 235, 236). The efficacy of molecular techniques is highly influenced by infection intensity, with higher sensitivity in high infection intensity indicated by the eggs per gram of feces measured through conventional fecal examination (220, 223, 224, 228, 233).

Besides application in the diagnosis of humans using feces, molecular techniques have also been developed in detection and differentiation of the larval stages (sporocysts, rediae, and cercariae) in snails (237, 238) and metacercariae in fish (234, 239–241). The application of molecular techniques in detection of larvae in snails could increase the sensitivity, when the larvae are still in the early development stages (e.g., sporocysts and rediae), and thus, they cannot be detected through the traditional cercarial shedding techniques (242).

Imaging

Imaging techniques are helpful to detect the damage and specific morbidity (e.g., cholangiocarcinoma), which, in turn, allow estimating the burden attributed to liver fluke infections (68, 72, 129, 243). With regard to liver fluke infections, imaging techniques include cholangiography, ultrasonography, computed tomography (CT), and magnetic resonance (MR) that are clinically used for auxiliary diagnosis (244, 245). The diagnosis is mainly based on two aspects, the imaging characteristics of adult worms and the related damage due to liver fluke infections. In cholangiography, many round filling defects of several millimeters in diameter within the bile ducts represent the adult worms (246, 247). It is challenging to detect worms in the bile ducts using ultrasonography. However, in heavy infection when worms parasitize the gallbladder, imaging appearances such as floating or dependent, discrete, non-shadowing echogenic foci in the lumen can be detected by ultrasonography (246, 248). Detection of flukes in CT is difficult (246). In MR, filling defects of the flukes could also be demonstrated, with small and irregular form (245).

Liver fluke infections can cause periductal fibrosis. Hence, dilatation of the small intrahepatic bile ducts with no or minimal dilatation of the large bile ducts is an important characteristic for diagnosis by the aforementioned imaging techniques, thus presenting as “too many intrahepatic ducts” sign in MR (245, 249). Appearances such as increased periductal echogenicity, diffuse dilatation of the intrahepatic bile ducts, floating echogenic foci in the gallbladder, and gallbladder distention in ultrasonography have been used for diagnosis of liver fluke infections (250, 251). Overall, the diagnostic accuracy, as indicated by the sensitivity and specificity, is rather low. The performance is influenced by the worm burden indicated by the infection intensity through counting eggs in feces, with low accuracy in light-intensity infection (250, 251). Additionally, the abnormalities in ultrasonography persist for a long time (up to several years) after clearance of flukes following drug treatment (252, 253).

TREATMENT

Praziquantel

As a broad-spectrum, safe, and efficacious anthelmintic against trematode and cestode infections, praziquantel is the only recommended drug for treatment of liver fluke infections (254). Praziquantel targets the cellular Ca²⁺ channels and pumps, which causes the disruption of Ca²⁺ homeostasis (254, 255). This then leads to the rapid contraction of the musculature and the formation of bleb and vacuole in tegument in parasites. WHO recommends a dose of 25 mg/kg three times daily for two to three consecutive days or a single dose of 40 mg/kg (256). High efficacy is shown with a dose of 25 mg/kg three times daily for two to three consecutive days. However, for ease of administration, a dose of 75 mg/kg divided into three doses of 25 mg/kg each in 1 day or a 50 mg/kg loading dose plus 25 mg/kg in 1 day is more frequently used (257). A recent systematic review determined a cure rate (proportion of patients who turned from egg positive to egg negative) of 79.8% for this treatment schedule in C. sinensis (split into three doses of 25 mg/kg each) and 93.8% in O. viverrini (50 mg/kg loading dose plus 25 mg/kg) (257). A single dose of 40 mg/kg resulted in cure rates of 32.4% in C. sinensis and 82.9% in O. viverrini. However, the corresponding egg reduction rates (percentage reduction of mean egg count) of schedules described above in both C. sinensis and O. viverrini were generally above 90% (258–263). The difference in cure rates observed in individual studies is predominantly attributed to the difference in pre-treatment infection intensity and diagnostic methods applied (257). In general, higher infection intensity is associated with lower cure rates. Additionally, gene polymorphism may affect treatment efficacy (264). It has also been reported in C. sinensis that levo-praziquantel (left isomer of racemic praziquantel) shows higher efficacy than praziquantel (265).

Albendazole

Albendazole, a broad-spectrum agent against nematodes and cestodes, is also efficacious against liver flukes (266). Albendazole blocks microtubule functions by inhibiting the polymerization of β-tubulin into microtubules, which subsequently leads to the decrease of glucose uptake and shortage of glycogen in parasites (266, 267). Indeed, a regimen of albendazole 10 mg/kg for 7 days is an alternative drug recommended for liver flukes by the US Centers for Disease Control and Prevention (19). The regimens of 8 mg/kg twice a day for 5 days and 10 mg/kg twice a day for 7 days both showed cure rates of 100% [39/39 (268) and 31/31 (269)], while the regimen of 5 mg/kg twice a day for 7 days showed a cure rate of 84.4% (27/32) in C. sinensis (269). However, in O. viverrini infection, albendazole 400 mg twice a day for 7 days only resulted in a cure rate of 33.3% (9/27), although the egg reduction rate was as high as 95.0% (270). Low dosage of albendazole (400 mg single dose) and mebendazole (400 or 500 mg) showed low efficacy in both C. sinensis and O. viverrini (271, 272).

Tribendimidine

Tribendimidine, a drug approved by the Chinese Food and Drug Administration for the treatment of soil-transmitted helminth infection, also shows high efficacy against liver flukes (273). The exact mechanism of tribendimidine against liver flukes is still unclear. It was speculated that the observed increase in lysophospholipids in C. sinensis triggers a destruction of the cell membrane and increases membrane permeability, resulting in an exposure to internal antigens, which subsequently are prone to an attack by host antibodies (274). The above mentioned systematic review determined a cure rate of 60.6% in C. sinensis and 89.8% in O. viverrini for a single dose of 400 mg in adults and 200 mg in children (257). The egg reduction rates were above 97% (260, 262, 272, 275). The lower cure rate observed in C. sinensis might be related to the high infection intensity. Additionally, in vivo and in vitro experiments demonstrated a potential effect of tribendimidine against O. felineus (276).

Treatment adverse events

Overall, adverse events of these drugs are mild and transient. Praziquantel produces somewhat more adverse events than albendazole and tribendimidine (262, 272, 275, 277). Dizziness/faintness, headache, and nausea are frequently reported following treatment with praziquantel, while abdominal pain, diarrhea, and vomiting are occasionally reported (262, 277, 278). Additionally, in spite of very few cases, anaphylactic reaction to praziquantel has also been documented (279, 280). The most frequently reported adverse events of albendazole are dizziness/faintness and headache, while abdominal pain and diarrhea are reported only occasionally (268, 269). Adverse events following tribendimidine treatment include vertigo, headache, nausea, fatigue, anxiety, and, occasionally, vomiting (260, 262, 272, 275).

CONTROL

Traditional strategy

The current strategy against liver fluke infections includes several major components, namely, chemotherapy for human beings, screening and management of cholangiocarcinoma attributed to liver flukes, behavioral change, control of animal reservoirs (e.g., cats and dogs), and environmental control.

Chemotherapy is the mainstay of control against human liver fluke infections, which is recommended by WHO and adopted in major endemic countries (Table 4) (256). The considerable global burden of clonorchiasis and opisthorchiasis calls for treatment to control morbidity. Moreover, this strategy decreases environmental contamination, which eventually helps to break transmission. Individual treatment, mass drug administration, or selective drug administration is applied according to prevalence levels in the community, but a standard for thresholds of prevalence level is not yet available. Individual treatment is indicated when the infection is ascertained by fecal examination (74, 281). Treatment is offered to the whole community in the frame of mass drug administration, while selective drug administration is provided only to specific populations at risk. The habitual ingestion of raw or undercooked freshwater fish is an important indicator for screening the population at risk (282–284). Highly endemic areas and adult males should be prioritized for chemotherapy (285). Usually, a regimen of 75 mg/kg praziquantel divided into three doses in 1 day or 40 mg/kg as single dose is used in mass or selective drug administration (74, 281, 286). In high-endemicity settings, mass drug administration once per year or selective drug administration twice per year is recommended, while in moderate endemicity, selective chemotherapy once per year is proposed (286, 287). Usually, 3 years of continuing chemotherapy significantly decreases the prevalence and infection intensity (286, 287). The major challenge of chemotherapy is sustainability because of rapid re-infection (73, 75). Re-infection is particularly high in adult males and those with past heavy-intensity infection (73, 75).

TABLE 4.

Measures against human liver fluke infections^a

Target	Strategy	Measures	Feasibility	Sustainability	Challenges
Human beings	Chemotherapy	Implementation of mass drug administration, selective drug administration for population at risk, or treatment for individuals with diagnosed infection	High	Low	High re-infection due to persistent ingestion of raw or undercooked freshwater fish
	Behavioral change	Abandoning the practice of ingesting raw or undercooked freshwater fish through education	Moderate (high in children and low in adults)	High	Adults especially men highly afflicted and reluctant to change behavior
	Management for cholangiocarcinoma	Screening of cholangiocarcinoma cases to control the morbidity	Moderate	Moderate	High dependence on the imaging techniques and lack of available cancer biomarkers
	Vaccine	Immunization with proteins, DNA, or other types of vaccines to produce antibodies against infection	NA	NA	Under development (preclinical stage)
Animals	Chemotherapy	Implementation of drug administration for animals (dogs, cats, pigs, etc.)	Moderate	Low	High re-infection due to persistent ingestion of raw or undercooked freshwater fish
	Vaccine	Immunization with proteins, DNA, or other types of vaccines to produce antibodies against infection	NA	NA	Under development (preclinical stage)
Environment	Feces management	Blocking the input of human and animal feces into water	Moderate (high in fish pond, low in large waterbodies (lakes, rivers and reservoirs))	Moderate	High cost and difficult to implement in large waterbodies
Snail	Physical control	Environmental modification of fish ponds to clear snail through the drainage of water and remove of bottom mud before the cultivation	Moderate	Low	Large waterbodies: difficult to clear; fish ponds: rapid re-introduction after clearance due to the residual snails or snails from external environment
	Chemical control	Using molluscicides	Low	Low	Large waterbodies: difficult to implement because of the aquatic habit of snails and potential toxicity to fish; fish ponds: potential toxicity to fish and rapid re-introduction due to the residual snails or snails from external environment
	Biological control	Control of snail density through biological competition or predation, e.g., other species of snails, fish, or animals; development of genetically modified snails with resistance	NA	NA	Preliminary study for biological competition and predation; theoretical exploration for genetically modified snails
Fish	Vaccine	Immunization with proteins, DNA, or other types of vaccines to produce antibodies against infection	NA	NA	Under laboratory experiments
	Chemotherapy	Drug administration for fish	NA	NA	Not yet reported

^a NA, not applicable.

Because of the strong association between liver fluke infection and cholangiocarcinoma that is responsible for high mortality, screening of cholangiocarcinoma in areas where liver flukes are endemic is important. At present, among the liver fluke endemic countries, large cholangiocarcinoma screening is only applied in Thailand, where a “cholangiocarcinoma screening and care program” covers the major endemic areas with opisthorchiasis (288–290). Individuals aged 40 years and above with infection or treatment history of O. viverrini or ingestion of raw or undercooked freshwater fish are included, undergoing ultrasound examination to detect liver and biliary abnormality. Suspected cases with cholangiocarcinoma are transferred to hospitals for further diagnosis.

Mathematical modeling indicates that chemotherapy alone cannot sustain the effectiveness (291). Thus, information, education, and communication (IEC) are needed to promote behavioral change to avoid ingestion of raw or undercooked freshwater fish. Individual education or education through community leaders demonstrates higher effectiveness compared to less targeted communication in entire communities (292, 293). However, many adults, especially males who are highly afflicted by liver flukes due to frequent ingestion of raw freshwater fish, might be hesitant to change their life style. They usually have some pieces of knowledge on liver flukes and refuse to the persuasion to giving up traditional raw-eating practice. Childhood is the key stage in sharpening the practice of raw freshwater fish consumption (60). Children are highly susceptible to education and behavioral change, which urge the adoption of IEC at school (294–296). However, because children are not the major afflicted population, the effectiveness in terms of prevalence in community is only shown after many years.

Animal reservoirs especially cats, dogs, and pigs play important roles in the transmission of liver flukes. A regimen of 40 mg/kg praziquantel shows high efficacy against liver fluke infections in animals (297, 298). However, as in humans, the prevalence rebounds quickly after chemotherapy (297). Management of human and animal feces decreases environmental contamination, including removal of simple toilets directly connected to fish ponds, disconnection of animals’ colony houses to water, and blocking of the access to waterbodies by animals through fences, which, in turn, successfully decrease the infection of liver fluke in fish (299–301).

Environmental control of fish ponds to clear snails through the drainage of water and removal of the bottom mud before the cultivation could also be useful, but snails could be rapidly re-introduced due to the residual snails or snails from external environment as well as the input of cercariae from surrounding freshwater bodies (302). Bayluscide (niclosamide) shows a high lethal effect in the laboratory, with a lethal concentration of 95% at <1 parts per million (ppm) (303). However, field trials demonstrated that the mortality is only 31.25% under 20 ppm. Furthermore, using molluscicides poses challenges because of the potential toxicity to fish. Aquaculture development promotes the use of large waterbodies, including rivers, lakes, and reservoirs in cultivation of fish, where the management of human and animal feces and control of snails is challenging and costly. Thus, integrated measures including two or more components of chemotherapy, IEC, and environmental management are gradually advocated and applied, which could increase the effectiveness and sustainability of control efforts (304, 305).

Potential future control strategies

Biological control techniques deserve to be explored. B. s. goniomphalos growth rates were hampered by the presence of another freshwater snail, Filopaludina martensi martensi (306). Of note, Radix auriclaria preys on young P. striatulus in laboratory (307). However, the potential introduction of other parasitic diseases should be monitored when introducing new species of snails. The effectiveness was also demonstrated by using black carp (Mylopharyngodon piceus) to control the density of freshwater snails (308). Additionally, ducks have been applied to control P. striatulus (309). Although ducks consumed a high number of P. striatulus at the beginning, they refused to eat again during a subsequent study (309).

In the early 2000s, crude antigens of adult worms, excretory-secretory antigens, and irradiated metacercariae of C. sinensis were tested in animals, which demonstrated some protective efficacy against re-infection (310, 311) (Table 5). Thus far, 10 proteins have been tested as potential vaccine candidates against infections with C. sinensis (312–322) and O. viverrini (323–326) (Table 5). Diverse types are applied, including protein vaccines, DNA vaccines, and Bacillus subtilis spore-based vaccines. However, these potential vaccines are still at early stage and currently tested in animal experiments. Thus far, only low or moderate efficacy was shown in animals in terms of worm reduction rate, varying between 15% and 70%. Fish vaccination is another potential control intervention. Enolase, cysteine protease, and paramyosin were tested in grass carps (Ctenopharyngodon idellus) against C. sinensis through oral administration as B. subtilis spore-based vaccine (327–329). None of 15 fish were infected when immunized by B. subtilis spore-based cysteine protease, while four were infected, nine were negative, and another two died in the control group (328). The number of metacercariae per gram of fish flesh was 7.2 in those immunized with B. subtilis spore-based paramyosin, lower than 13.2 in the control group (329). Although praziquantel has been applied in fish cultivation, no report has been documented for liver fluke infections (330).

TABLE 5.

Tested vaccines against human liver fluke infections in animals^h

Vaccine type	Year of publication	Protein	Animal species	Immunization route	Worm reduction rate (%)	Egg reduction rate (%)
C. sinensis
Protein vaccine (310)	2000	Crude antigens of adult worm	Albino rats	Subcutaneously and intravenously	23.4	NK
Protein vaccine (310)	2000	Excretory-secretory antigens	Albino rats	Subcutaneously and intravenously	35.6	NK
Irradiated metacercariae (311)	2005	Irradiated metacercariae	Sprague-Dawley rats	Oral administration	NKa	NK
DNA vaccine (pcDNA3.1) (312)	2006	Cysteine proteinase	Sprague-Dawley rats	Subcutaneously	31.5	15.7
DNA vaccine (pcDNA3.1) (313)	2006	Fatty acid-binding protein	Sprague-Dawley rats	Subcutaneously	40.9	27.5
Bacillus subtilis spore-based vaccine (314)	2008	Tegumental protein 22.3 kDa	Sprague-Dawley rats	Oral administration	44.7	30.4
Virus-like particles vaccine (315)	2017	Tegumental protein 22.3 kDa	Sprague-Dawley rats	Intranasally	70.2	NK
Protein vaccine (316)	2010	Rho GTPase	Sprague-Dawley rats	Subcutaneously and intraperitoneally	60.5	68.8
Protein vaccine (317)	2012	The 14-3-3 epsilon	Sprague-Dawley rats	Subcutaneously	45.4	37.9
Protein vaccine (318)	2012	Paramyosin	Sprague-Dawley rats	Subcutaneously	54.3	50.9
DNA vaccine (pcDNA3.1) (318)	2012	Paramyosin	Sprague-Dawley rats	Intramuscularly	36.1	38.8
Bacillus subtilis spore-based vaccine (319)	2018	Paramyosin	BALB/c mice	Oral administration	NKb	48.3
Bacillus subtilis spore-based vaccine (319)	2018	Paramyosin	BALB/c mice	Intraperitoneally	NKb	51.2
Protein vaccine (320)	2014	Enolase	Sprague-Dawley rats	Subcutaneously	56.3	NK
Protein vaccine (320)	2014	Enolase	Sprague-Dawley rats	Intraperitoneally	15.4	NK
DNA vaccine (pcDNA3.1) (320)	2014	Enolase	Sprague-Dawley rats	Intramuscularly	37.4	NK
Bacillus subtilis spore-based vaccine (320)	2014	Enolase	Sprague-Dawley rats	Oral administration	61.1	80.7
Protein vaccine (321)	2014	Cathepsin B cysteine protease (CB2)	Sprague-Dawley rats	Subcutaneously	41.0	33.6
Protein vaccine (321)	2014	Cathepsin B cysteine protease (CB3)	Sprague-Dawley rats	Subcutaneously	67.2	57.7
Protein vaccine (322)	2020	CsAg17 protein	FVB mice	Intraperitoneally	64.0	NK
DNA vaccine (pcDNA3.1) (322)	2020	CsAg17 protein	FVB mice	Intramuscularly	69.0	NK
O. viverrini
Protein vaccine (323)c	2019	O. viverrini tetraspanin-2d	Syrian golden hamsters	Intraperitoneally	34.0	41.0
Protein vaccine (324)e	2019	O. viverrini tetraspanin-2d	Syrian golden hamsters	Intraperitoneally	31.1	38.0
Bacillus subtilis spore-based vaccine (325)	2021	O. viverrini tetraspanin-2d	Syrian golden hamsters	Oral administration	56.0	56.0
Protein vaccine (326)	2020	Chimeric proteinf	Syrian golden hamsters	Intraperitoneally	27.2	NKg

^a The worm count decreased compared to that in control, but no data provided in detail.

^b Failed to detect worm.

^c Produced in Escherichia coli.

^d The large extracellular loop of O. viverrini tetraspanin-2.

^e Produced in Pichia pastoris.

^f The large extracellular loop of O. viverrini tetraspanin-2 and leucine aminopeptidase.

^g No statistical significance but no data provided in detail.

^h Data for C. sinensis between the years of 2006 and 2014 were also summarized in reference (12); NK, not known.

Potential value of multi-omics in control

Considerable advances have been made in omics disciplines, including genomics, transcriptomics, and proteomics for liver flukes in the past decade (331, 332). The genomes of all three species of liver flukes have been annotated, which increases the knowledge on biology, pathogenicity, and parasite-host interaction (333–336). For example, genomic comparison of C. sinensis individuals from five different populations in the People’s Republic of China and one from the far east of Russia showed significant genetic variation between the individuals from “southern” and “northern” geographic regions (337). Biologically distinct variants are likely present in C. sinensis, which may lead to the difference in epidemiology, pathogenicity, and even chemotherapy responsiveness (337). Not only the transcriptomes of adult worms have been explored for the three species of liver flukes, but also the transcriptomes of juvenile (of O. viverrini) and metacercaria (of O. felineus) have been analyzed and compared to those of the adults, which increase the knowledge on the developmental transition between different stages and in different hosts (338–340). Furthermore, transcriptomes from multiple C. sinensis tissues (sucker, muscle, ovary, and testis) have also been compared (341). Attention is being paid to excretory-secretory products and tegument in proteomic studies because they are the interface of parasite-host interactions and thus show high potentiality as diagnostic, drug, and vaccine targets (342, 343).

Current diagnosis of cholangiocarcinoma is highly dependent on the imaging techniques, and thus, cancer biomarkers are expected for screening, early detection, diagnosis, prognosis, and follow-up in liver fluke-related cholangiocarcinoma (344). Exome sequencing demonstrated that TP53, KRAS, and SMAD4 show the most frequent mutations in O. viverrini-related cholangiocarcinoma (345). Further comparison on O. viverrini-related and non-O. viverrini-related cholangiocarcinoma revealed differential mutational patterns, with frequent mutation of TP53 in the former (346). Based on multi-omics approach, liver fluke-related cholangiocarcinoma cases are enriched in ERBB2 amplifications and TP53 mutations compared to those non-liver fluke-related cholangiocarcinoma cases (347). Gene expression profiles are also different between O. viverrini-related and non-O. viverrini-related cholangiocarcinoma, exhibiting increased expression of genes involved in xenobiotic metabolism in the former (348). Elevated expression of 14-3-3 eta is found in O. viverrini-infected hamsters through proteomics analysis, and the percentage of high expression of 14-3-3 eta in cholangiocarcinoma reached 84.2% (187 out of 222 cases), while it was not detected in hepatocellular carcinoma and healthy liver tissues (349). Proteomic analysis in animal models also showed the overexpression of plasma orosomucoid 2, and the sensitivity and specificity in using elevated orosomucoid 2 to discriminate cholangiocarcinoma cases from healthy controls were 92.9% (65 out of 70 cases) and 73.7% (350). Exostosin 1 in plasma is also found to be upregulated through proteomic analysis in animal models, and the upregulation was detected in 89.7% of cholangiocarcinoma cases (61 out of 68) (351). However, further studies are needed to explore the use of these potential cancer biomarkers in clinical settings (344).

The extreme low infection rate of snail hosts by liver flukes shows an amplification role in transmission and also indicates potential defensive mechanisms in snails (331). Based on the transcriptome of B. s. goniomphalos (352), differences in gene transcription between O. viverrini-infected and uninfected B. s. goniomphalos are captured (353). Compared to the uninfected snails, transcripts encoding distinct proteases are highly downregulated in infected snails, while those encoding heat-shock proteins and actins are significantly upregulated (353). Proteomic analysis also demonstrates the expression changes in O. viverrini-infected B. s. goniomphalos, with a downregulation in the expression of oxidoreductases and catalytic enzymes and upregulation of stress-related and motor proteins (354). Furthermore, compared to the headfoot tissues, the differences of protein expression after O. viverrini infection are more obvious in the body of the snails (354). Additionally, proteomic analysis on the hemolymph of B. s. goniomphalos demonstrates the upregulation of proteins related to immune response in both hemocyte and plasma in infected snails (355). These pieces of evidence probably point to the existence of differences in susceptibility of snails to liver flukes (331). Further understanding and decoding on the differences may promote the development and application of genetically modified snails for control of liver fluke infections (331).

CONCLUDING REMARKS AND FUTURE PERSPECTIVES

High endemicity of clonorchiasis and opisthorchiasis is driven by behavioral, social, economic, and cultural factors. Although our knowledge on these liver flukes and available techniques tackling the infections have increased significantly over the past several decades, gaps still exist, which hinder control, let alone elimination (Box 1) (12, 356–358). Nowadays, national surveys are only regularly implemented in the People’s Republic of China, South Korea, and Thailand. Data pertaining to O. viverrini infection in Lao People’s Democratic Republic, Cambodia, southern Vietnam, and Myanmar are inadequate, while data of O. felineus infection are scarce. Moreover, not only prevalence surveys in humans but also data on animal reservoirs and intermediate hosts are needed. Additionally, environmental, social, economic, behavioral, and cultural factors as well as aquaculture influencing transmission should be researched simultaneously, which then could be integrated with prevalence data to establish high-resolution maps through geostatistic techniques (24–26).

Box 1: Proposed key research areas for liver fluke infections.

1. To update the global epidemiology of liver fluke infections

   • Implementation of well-designed surveys and establishment of surveillance-response systems in endemic areas.

   • Mapping the definitive distribution of the three main species of liver flukes based on molecular epidemiological surveys.

2. To disentangle the transmission

   • Quantifying the roles of different species of hosts in transmission, which may vary by prevalence, ecology, and species of liver flukes.

   • Illuminating the circulation model between human and animal hosts (i.e., full cycle, partial cycle, or separate cycle).

   • Exploring the potential of interspecies transmission among the three key liver fluke species.

   • Predicting the impact of climate change on transmission of snails and subsequent change in distribution of liver flukes.

3. To estimate the global burden

   • Besides liver and biliary system, the relevance of liver flukes to introduce other pathology (e.g., kidney and heart) needs to be determined and quantified.

   • The carcinogenic potential of O. felineus in cholangiocarcinoma, as well as the association between liver flukes and primary hepatocellular carcinoma, needs to be elucidated.

   • The association of liver fluke infections and allergy needs scientific inquiry.

   • The roles of other factors in morbidity, especially co-infection with hepatitis virus, and H. pylori on carcinogenicity need to be explored.

4. To develop new diagnostic tools

   • Establishment of machine learning for fecal examination and development of products.

   • Development of sensitive and specific immunological and molecular diagnostic tools.

   • Development of eDNA techniques for risk evaluation, not only for liver flukes but also for intermediate hosts, especially snails.

5. To develop new drugs

   • Establishing a clinical development plan for tribendimidine against C. sinensis and O. viverrini, and implementing regulatory clinical trials against O. felineus.

   • Initiating randomized controlled trials to verify the efficacy of albendazole using praziquantel as comparator.

   • Exploration on the efficacy of drug combination against liver flukes.

6. To explore sustainable control strategies

   • Optimization and combination of traditional available measures, in a cost-effective manner.

   • Further studies on potential new techniques, including fish vaccine and biological control through species competition.

   • Further mining, validation, and application of the multi-omics findings, for the diagnosis in infection and morbidity (especially cholangiocarcinoma), and discovery of new drug and vaccine targets, as well as the potential development of genetically resistant snails.

   • Validation and application of a One Health approach integrating measurements targeting human beings, animal reservoirs, two intermediate hosts, and environments for sustainable control.

Although the three species of liver flukes have separate distribution limits, this is not based on systematic evidence from molecular epidemiology. It is still debated whether C. sinensis is endemic in Thailand. The diversity of species in transmitting liver flukes challenges the control, but the roles of different species are unclear. It is of crucial importance to illuminate the precise transmission in humans and the role of animal reservoirs. Currently, only little evidence from genetic research is available, which does not allow to draw a clear conclusion on the interplay between humans and animals, which urges further studies (359). The differential distribution of snail hosts is crucial in the distribution of three species of liver flukes. However, interspecies transmission cannot be completely excluded. Furthermore, climate change is likely to exert an impact on liver flukes and intermediate hosts, thus influencing on transmission.

Although the association of liver fluke infections and damage in liver and biliary systems is well recognized, the global burden of liver fluke infections might still be underestimated because the pathology in other organs and systems is not clear, and the role in allergy remains to be investigated. Whether O. felineus is a carcinogen needs more research. Although co-infection with other pathogens (e.g., hepatitis virus and H. pylori) is common, their interaction is little studied.

Important gaps also still remain in diagnostic and treatment tools. Novel machine learning techniques could free technicians from tedious laboratory examination, but this line of scientific inquiry needs further development and rigorous validation. Thus far, many immunological and molecular techniques have only been tested in the research setting. Samples from infected cases and non-infected individuals are often from different sources, and few studies compared different diagnostic techniques and/or products side by side. Guidelines should be established how to quantify the intensity in immunological and molecular techniques. Moreover, environmental DNA (eDNA) is a highly neglected area in liver flukes (360). This technique will strengthen the risk evaluation and prediction in the future through the detection of liver flukes in water and the existence of suitable intermediate hosts.

A clinical development plan should be discussed for tribendimidine, including the implementation of regulatory trials in O. felineus. Because of the significant differences in the efficacy of albendazole against C. sinensis and O. viverrini infections, further trials are needed to verify the efficacy of albendazole. Drug combination is neglected in the treatment of liver fluke infections, which is also expected to be strengthened in future because of potentially synergistic effects owing to distinct mechanisms from different drugs.

How to combine and optimize the different available measures, including chemotherapy, behavioral change, and environmental control, will determine the cost and effectiveness and help to identify the ideal measures for short-term morbidity control, long-term transmission control, and even interruption. Fish vaccines deserve to be studied in greater detail, while biological control through snail competition potentially increases the sustainability of control of liver fluke infections. There exists high potential to mine and utilize the findings of multi-omics in combating liver flukes. Especially, sensitive and specific cancer biomarkers for liver fluke-related cholangiocarcinoma might be a game changer in screening and early diagnosis. Additionally, further validation of the snail resistance and subsequent decoding of the mechanisms may benefit the development and application of genetically modified snails for sustainable control and even elimination. The One Health approach emphasizes the comprehensive collaboration on health at the human-animal-environment interface (361). The WHO roadmap pointed out the importance that the WHO, the Food and Agriculture Organization of the United Nations, and the World Organization for Animal Health (formerly OIE) collaborate to promote the One Health approach through multisectoral action in clonorchiasis and opisthorchiasis (2). On one hand, mathematical modeling could not only predict the effectiveness and cost but also present the conditions for transmission control and interruption, which could guide the development on sustainable control strategies (362). On the other hand, validation is needed before the application of special One Health approaches, which should take considerations of the differences in transmission determinants in different settings.

During the past decades, considerable progress has been made for many parasitic diseases, some of which are nowadays even eliminated or targeted for elimination. However, control of liver fluke infections lags behind. In some endemic areas, the prevalence even increases due to the availability of and accessibility to freshwater fish. As high neglected diseases, concerted efforts are needed with cooperation and partnership between government departments, the scientific community, and the public and private sectors.

Biographies

Men-Bao Qian heads the Department of Soil-transmitted and Food-borne Parasitic Diseases, in the National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention (Chinese Center for Tropical Diseases Research). He studied in Chinese Center for Disease Control and Prevention for his MSc in Pathogen Biology between 2006 and 2009, and then for PhD in Epidemiology and Health Statistics during 2016 and 2019. His team is in charge of the design and implementation of national surveys and surveillance, as well as the development and evaluation of diagnosis and treatment tools, and the exploration of control and elimination strategies, for soil-transmitted helminthiasis and food-borne parasitic diseases.

Jennifer Keiser heads the Helminth Drug Development Unit at the Swiss Tropical and Public Health Institute (Swiss TPH). She is an Associate Professor of Neglected Tropical Diseases at the Faculty of Science of the University of Basel. Prof. Keiser received a PhD in Zoology from the University of Basel in 1999 and was a Postdoctoral Fellow at Princeton University, NJ, USA from 2000 to 2003. Research objectives of her team include in vitro and in vivo evaluation of biological activities of chemical compounds, assay development, preclinical studies such as pharmacokinetics and metabolism, and clinical trials in helminthiasis-endemic countries, particularly in Côte d’Ivoire, Lao People’s Democratic Republic, and Tanzania.

Jürg Utzinger is the Director of the Swiss Tropical and Public Health Institute (Swiss TPH) and Professor of Epidemiology at the University of Basel. He holds an MSc in Environmental Science from ETH Zurich and a PhD in Epidemiology from the University of Basel. He pursued several years of postdoctoral research in demography and epidemiology at Princeton University in the United States of America. Utzinger’s research and teaching interests pertain to the epidemiology and integrated control of neglected tropical diseases and health impact assessments of large footprint projects in low- and middle-income countries. He is engaged in trans-national global health research consortia with ongoing projects in the People’s Republic of China, Côte d’Ivoire, South Africa and Tanzania.

Xiao-Nong Zhou has been Director of the National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention (Chinese Center for Tropical Diseases Research) and the World Health Organization (WHO) Collaborating Centre for Tropical Diseases since 2010. He is also serving as Vice Dean, School of Global Health, Chinese Center for Tropical Diseases Research, Shanghai Jiao Tong University School of Medicine. Prof. Zhou, graduated with a PhD on biology from the University of Copenhagen in 1994, is a leading expert in the research and control of tropical diseases, with over 40 years’ experience on control of parasitic diseases, and his professional works are across the fields of ecology, population biology, and epidemiology of tropical diseases.