Opisthorchis viverrini

In brief

What it is, and how it gets into a person

Opisthorchis viverrini is a trematode — a parasitic flatworm, in the family Opisthorchiidae — whose adult stage lives in the bile ducts, the branching tubes that carry bile from the liver to the intestine. It is one of three closely related liver flukes of medical importance: Clonorchis sinensis in eastern Asia, O. viverrini in southeastern Asia, and Opisthorchis felineus in eastern Europe and Siberia ¹. All three share one life cycle involving humans or animal reservoirs and two intermediate hosts, and all three are associated with liver and biliary complications, most importantly cholangiocarcinoma, which is cancer of the cells lining the bile ducts ¹. Together they infect roughly 25 million people ⁵.

The cycle is fixed and every step is required. Eggs leave a definitive host — a person, a cat or a dog — in feces, reach fresh water, and are eaten by a small freshwater snail; in this species the snail is of the genus Bithynia ². Inside the snail the parasite multiplies and emerges as a swimming larva, the cercaria, which burrows into a freshwater fish — usually a cyprinid, a fish of the carp and minnow family — and encysts in its flesh as a metacercaria, the dormant form a person swallows ². Adults then attach to the bile duct lining and feed there for as long as 10 to 30 years ². Direct field measurement in central Vietnam shows how narrow the snail step is: of 12,000 snails from six families, only the two Bithynia species shed cercariae, at 0.86% and 0.14%, while metacercariae were found in 10 of 12 fish species, highest in Carassius auratus at 74.0% ⁴. Those ten species spanned both cyprinid and non-cyprinid families, so the fish step is less narrow than the snail step ⁴. Cats matter more than dogs. A 2026 meta-analysis of surveys across the Mekong region put infection at 11.97% in cats against 1.89% in dogs, and cat infection tracked human infection closely from place to place, so treating only people leaves a reservoir in the village ²⁴.

The fluke is built for its address. Its draft genome, 634.5 million base pairs — base pairs being the paired chemical letters that DNA is written in — shows metabolic pathways adapted to a lipid-rich diet from bile and from bile duct cells, and its secretions include proteins that directly drive host cell proliferation ¹⁶. Nor is “the species” one thing: molecular work indicates a complex of at least two cryptic species in Thailand and Laos — populations that look identical but are genetically distinct enough to count as separate species — divisible into six genetic groups that correlate with five river wetland systems, and the groups differ measurably in how well they establish in a host and how many eggs they lay ¹⁷. Its haploid set of six chromosomes (diploid 2n = 12) has been measured chromosome by chromosome, from 0.84 to 2.84 µm ²⁵.

Infection is a matter of cuisine, not of hygiene in the usual sense. The dishes that transmit it are raw freshwater-fish preparations such as koi pla, a raw minced fish salad, and lightly fermented fish such as pla som ^2,26. The habit is strongly gendered and culturally held: in two northeastern Thai villages men ate koi pla and pla som on 64.5 and 40.4 days a year — one figure per village, counting both dishes together — against 4.1 and 4.3 days for women, and taking part in sharing raw-fish dishes cut the odds that a person would be willing to give them up (odds ratio 0.19) ²¹. In four villages in Kalasin province, 76.9% gave deeply embedded cultural practice as a reason for continuing, about 30% said they would not stop even knowing the consequences, and only 15.3% correctly identified raw fish as the source of the infection while 84.2% believed other raw foods could transmit it ²⁶.

The whole history, including what was overturned

The species was named by Poirier in 1886, the date carried in its formal name ⁴³, and the discovery history of the opisthorchiid flukes goes back more than 200 years ⁴⁴. Its modern study in Thailand began in the 1950s, with Sadun's field surveys and the first treatment trials, of quinacrine and chloroquine, in 1955 ^45,46; the life cycle was worked out and compared with that of its Russian cousin O. felineus by 1965 ⁴⁷. Control began early — a small helminthiasis control program in high-risk areas in 1950 — but the drug it came to rely on arrived only in 1980, when praziquantel cleared eggs from all 49 patients in the first Thai trial, and a single 40 mg/kg dose was recommended for control the following year ^48,49. The national liver fluke program came to rest on three things: stool examination with praziquantel for those positive, health education promoting cooked fish, and sanitation to break transmission; reported positivity fell from 63.6% in 1984–87 to 9.4% in 2001 ⁵⁰.

The cancer link was built in two directions at once. Clinically, the pattern was visible from the registry: in northeast Thailand about a third of the population was infected, and bile duct cancer was among the leading causes of death in the region ⁵¹. Experimentally, Syrian hamsters given the fluke plus a nitrosamine developed cholangiocarcinoma while neither agent alone did it ^37,39,52. Epidemiologically, a case-control study of 103 patients found an odds ratio of 5.0 for past or present infection, with at least two-thirds of cases attributable — and no association with hepatitis B, with aflatoxin measured as serum albumin adducts, or with any particular dietary pattern — though regular betel-nut use, mainly among women, carried an odds ratio of 6.4, which the authors linked to nitrosamine exposure ⁵³. A population survey of 12,311 adults then found the dose-response: among the most heavily infected, the prevalence odds ratio for ultrasound-detected cancer was 14.1 ⁵⁴.

That was the evidence the International Agency for Research on Cancer (IARC), the World Health Organization's cancer agency, had when its working group met in Lyon in June 1994 and placed O. viverrini in Group 1, carcinogenic to humans ⁷. A second working group reviewed the biological agents in 2009, announced its conclusions that April, and published the 441-page monograph in 2012 — which is why the dates for these classifications are often written as 2009/2012 ^8,9. The 1994 assignment stood, and the fluke remains one of the 13 infectious agents in Group 1, and one of the 12 whose cancer burden IARC quantified in its 2026 analysis ¹⁸.

Then the mechanism work turned from description to intervention. Granulin, the fluke's own growth factor, was identified in its secretions in 2009, silenced by RNA interference in 2014, and in 2019 knocked out of living flukes with CRISPR/Cas9 — one of the first two programmed gene edits in a parasitic flatworm, published on the same day, 15 January 2019, and in the same journal as a knockout of the omega-1 gene of the blood fluke Schistosoma mansoni ^19,31,32,55. In 2022 the same knockout was run with dietary nitrosamine to a cancer endpoint: fibrosis, cell proliferation and p53 mutation fell markedly, and the edited worms produced fewer and lower-grade cancers — 7 of 13 hamsters against 10 of 12 — a difference not statistically secure at these group sizes ²⁰. Meanwhile the epidemiology reversed in two places. Khon Kaen's cholangiocarcinoma incidence, long cited as the highest in the world for this cancer, is now falling steadily and by a birth-cohort pattern ^11,12. And the belief that killing the worm must lower cancer risk ran into human data pointing the other way, which is still not resolved ^14,56.

Where it is now, and how many people have it

The honest summary is that the geographic center has moved and the count depends on the test. O. viverrini is endemic in Thailand, Lao PDR — the Lao People's Democratic Republic, Laos — Cambodia and central and southern Vietnam, with more than 12 million people infected, while C. sinensis accounts for an estimated 3.5 to 5 million in northern Vietnam and Guangxi ⁶. Thailand's measured prevalence fell from over 15% in 1996 to 2.2% in 2019 after three decades of national campaigns, while high prevalence persists in Lao PDR and central Vietnam, and emerging figures up to 47.5% have been reported in Cambodia over the past decade — which the authors suggest may reflect more studies being done rather than a real rise ⁶. The high-risk area has, in their words, shifted from northeastern Thailand to Laos, Cambodia, Vietnam, Myanmar and Guangxi ⁶. Current surveys in Laos bear that shift out. In 2026, stool examination of 1,530 randomly chosen adults in 81 villages of Khammouane province found the liver fluke in 56.9%, with minute intestinal flukes in 22.7%, and better water and sanitation made no difference to either ⁶⁵; in three villages of Champasak, 51.7% were positive by stool and 65.3% by an antibody strip, for an estimated true prevalence of 66.6% ⁶⁶.

Thailand's own national survey makes the texture of what remains clear. In 2019 the Ministry of Public Health examined 16,187 stool samples from all 76 provinces: overall helminth prevalence 9.79%, with hookworm first at 4.47% and this fluke second at 2.2%, mean intensity 120.9 eggs per gram, and 99.1% of fluke infections classed as low intensity ⁶⁷. Light infection is now the norm, and light infection is what stool microscopy is worst at finding.

That is why the 2025 urine antigen survey matters so much. Across 23 sub-districts in 21 provinces, 19,465 urine samples and 18,929 stool samples from the same people gave 50.3% prevalence by antigen against 12.2% by stool; measured against the combined reference, sensitivity was 91.6% for the antigen assay and 21.9% for stool examination; and the ratio between the two estimates was largest in the provinces with the lowest egg counts ¹³. That comparison treated anyone positive on either test as infected, so it cannot measure how often the antigen assay is positive in an uninfected person, and it flatters the assay's sensitivity ¹³. Smaller studies point the same way: in 246 people in Sakon Nakhon in 2025, microscopy found 6.1% — every case very light — and ITS2 PCR found 19.9% ⁶⁸; and in 307 rural participants, stool concentration found 16% while antibody tests were positive in 26.1% and 31.6% ⁶⁹.

Burden figures should be read with that in mind. The cancer end of the problem is concentrated: in registry data from 38 countries for 1993 to 2012, Thailand had the second highest rate of intrahepatic cholangiocarcinoma, 2.19 per 100,000, behind South Korea at 2.80, and the largest rise in extrahepatic rates of any country, 8.8% a year ⁷⁰. The Khon Kaen registry recorded 18,589 liver cancers between 1985 and 2009, of which 58% were cholangiocarcinoma rather than hepatocellular carcinoma — cancer of the liver cells themselves, and the inverse of the global pattern — with an age-standardized rate of 44.3 per 100,000 in men (95% CI 38.9–49.7) and 17.6 in women (14.5–20.7) ¹¹. The national figure is twenty times lower than Khon Kaen's for two reasons: it pools registries across a whole country, and most liver cancers in the northeast are diagnosed without tissue, so they cannot be assigned to a subtype at all — only 10.8% of Khon Kaen's liver cancers were verified by cytology or histology ¹¹. IARC's 2026 global analysis attributes about 3,800 new cases of intrahepatic cholangiocarcinoma in 2024 to O. viverrini and C. sinensis together, 2% of all intrahepatic cholangiocarcinoma worldwide, and counts no tumors of the ducts outside the liver ¹⁸; that is a figure set by method rather than a count, and Khon Kaen province alone recorded about 430 cholangiocarcinomas a year across those 25 years (our arithmetic from the registry totals) ¹¹. In southern Lao PDR, among 3,400 people in highly endemic villages in 2024, 88.3% were infected and 7.2% (95% CI 5.4–9.6) had ultrasound-suspected cholangiocarcinoma, an imaging suspicion rather than a biopsy diagnosis ⁷¹. For scale, the Thai screening program, which confirms its cancers by tissue, found cancer in 0.11% of 751,061 people screened ⁷²; a suspicion rate sixty times higher is reason to expect that many of those suspicions would not be confirmed.

What it does inside a bile duct

For most people, for years, nothing. The infection is asymptomatic until the worm burden is high, and then presents as chronic cholangitis — inflammation of the bile ducts; complications accumulate over time as gallstones, cholangitis, liver abscess and pancreatitis, and after decades as cholangiocarcinoma ². The pathology the fluke leaves while it feeds is described consistently: chronic inflammation, thickening of the duct lining (epithelial hyperplasia), scarring around the ducts (periductal fibrosis), and granulomas around trapped eggs ².

Periductal fibrosis is the change that matters, because it is visible on ultrasound and it marks risk. Of 3,359 egg-positive people scanned in Khon Kaen province, 23.6% had advanced periductal fibrosis, more often men, and the risk rose with infection intensity; eleven suspected cancers were seen in the same series ²². In the much larger screening program, 15.35% of 751,061 participants had fibrosis of some grade, and the grade mattered: fibrosis of the segmental ducts carried an adjusted odds ratio for cancer of 1.4 and of the main duct 2.52, while fibrosis confined to the peripheral ducts carried none ⁷².

There is a measurable inflammatory signature in blood. Plasma interleukin-6 — a cytokine, one of the chemical messages immune cells use — was 58 times higher in people with advanced fibrosis than in matched controls and 221 times higher in people with bile duct cancer; those in the highest quarter had 19 times the odds of advanced fibrosis and 150 times the odds of cancer compared with people with no detectable interleukin-6 — measured at the same moment, so these figures describe who has the disease, not who will get it ²³. The authors read this as evidence that one process drives both the fibrosis and the tumor. These are associations measured at one time point, not a demonstration that interleukin-6 causes either.

Other conditions add to the risk. In 263,776 people screened between 2013 and 2019, 32.4% reported past fluke infection and 8.2% diabetes; cancer was found in 0.47% of the infected, 0.59% of the diabetic and 0.73% of those with both, and the adjusted odds ratio for the combination was 2.36 (95% CI 1.74–3.21) against neither, a combined effect the authors describe as greater than that of either condition alone; they report no test of whether the two multiply rather than add ⁷⁵. The cancer, when it comes, is usually found late: five-year relative survival in the Khon Kaen registry was 10.9% (95% CI 10.3–11.6) ¹², and one review describes five-year survival under 5% for unresectable disease ².

The mechanisms, and the class of evidence each rests on

The standard account names three intertwined mechanisms: direct damage by the adult worms, immunopathology and oxidative stress from chronic inflammation, and the mitogenic and anti-apoptotic effects of proteins the parasite secretes — with dietary nitrosamines as an outside cofactor ². That account is sound, but the three strands are not equally supported, and the honest version separates them by evidence class.

The mechanical strand is the weakest in direct evidence. There is no measurement isolating lesion burden caused by sucker attachment and grazing from the inflammation that follows. The nearest thing is an experiment with no parasite at all: hamsters initiated with one 20 mg/kg dose of dimethylnitrosamine and then given complete surgical ligation of a bile duct developed 60.9% cholangiofibrosis, 21.7% mucous cystadenomas and 39.1% cholangiocarcinoma at week 40, where the chemical alone produced cholangiofibrosis in 5% and no tumors at all ²⁸. That shows obstruction and the regenerative proliferation it forces can substitute for the worm; it does not quantify the worm's own mechanical contribution.

The inflammation strand is well measured in animals and partly in people. In hamsters, repeated infection accelerated inflammatory infiltration and raised 8-nitroguanine and 8-oxo-7,8-dihydro-2'-deoxyguanosine in the duct lining in the order three infections above two above one, with the lining cells themselves expressing inducible nitric oxide synthase by day 90 ²⁹. Praziquantel given for one week, which did not kill every worm at days 14 and 30, nonetheless almost abolished both lesions and lowered nuclear factor kappa B and the enzyme — which the authors read as an anti-inflammatory effect of the drug beyond killing worms ³⁰. In people, urinary 8-oxodG — the shorthand for the oxidative DNA lesion named in full above — tracks fibrosis and falls after treatment only in those without established fibrosis ⁷⁶.

The secreted-protein strand is now the best evidenced, because it is the only one that has been tested by removing a single molecule. Ov-GRN-1, the fluke's granulin, drives proliferation of mouse fibroblasts at nanomolar concentrations, and antibodies to it inhibit most of the proliferative activity of the whole secretion mix ³¹; laboratory-made Ov-GRN-1 at low nanomolar levels drives human cholangiocytes to over-proliferate ³²; silencing the gene by RNA interference cut proliferation of cholangiocytes by 25% and of a cholangiocarcinoma line by 92% in co-culture ³²; and knocking the gene out reduced fibrosis, cell proliferation and p53 mutation in infected hamsters, with fewer and lower-grade cancers in groups too small to make that last difference secure ^19,20. Alongside it, thioredoxin peroxidase protects the worm from the radicals inflammation throws at it and was detected in bile and in bile ducts too small for the worm to enter ³⁴; the fluke's thioredoxin blocked oxidative-stress apoptosis in cholangiocytes and bound apoptosis signal-regulating kinase 1 ³⁵; and secreted extracellular vesicles, found in bile from infected hamsters and infected people, drove proliferation and interleukin-6 release ³⁶. A fifth strand sits outside the worm's own biology: the fluke harbors Helicobacter pylori, and hamsters co-infected with the cagA-positive strain had higher inflammation, earlier dysplasia and a higher proliferation index ⁴².

Two chemical pathways deserve separate treatment because they are often stated more strongly than the data allow. Oxysterols — oxidized derivatives of cholesterol — have been detected in fluke extracts, including derivatives compatible with an estrogen core and with oxidation to an estradiol-2,3-quinone ⁷⁷, and five were identified in livers of infected hamsters, of which two were significantly higher in animals with cancer and both induced DNA adducts and apoptosis in cultured human cholangiocytes ⁷⁸. The argument that these are initiators of the cancer is a hypothesis with supporting in vitro and animal chemistry, not a demonstrated human mechanism ⁷⁹. The nitrosamine pathway is firm in the hamster ^37,38,39 and has human measurements at both ends — raised endogenous nitrosation in antibody-positive people living in high-risk areas, although in that same study area-level cancer incidence did not correlate with any of the nitrosation measures across five contrasting areas of 20 people each ⁴⁰, and nitrosamines measured directly in fermented and dried fish sampled from four major regions of Thailand ^40,41 — but no study has linked a measured dietary nitrosamine intake in an individual to that individual's cancer.

The gene that was cut out of the worm

This is the most direct genetic test in the parasite-cancer literature, and a small one. It is also routinely described wrongly, so it is worth being exact. The gene that was edited belonged to the fluke, not to the host. Ov-grn-1 encodes Ov-GRN-1, a secreted single-domain homolog of human granulin, abundant in the fluke's excretory-secretory products ³¹. Nothing in the hamster's genome was touched. A separate paper by the same group did knock out the host gene, human progranulin, but only in a cultured human bile duct cell line, in order to read the fluke protein's effect without the host's own version in the background — that paper is almost certainly the source of the confusion ⁸⁰.

The three experiments say different things and should not be merged. The 2019 paper measured precancerous change with no carcinogen: edited flukes colonized bile ducts and matured, but biliary epithelial thickening at day 14 was 145% of uninfected against about 500% for wild-type infection, and collagen was 28% lower at day 60 (p ≤ 0.001) ¹⁹. The 2022 paper added dietary nitrosamine and ran to 24 weeks, and it is the one that measured cancer: 10 of 12 control hamsters developed cholangiocarcinoma, 8 of them high-grade, against 7 of 13 in the knockout group, 6 high-grade; mutant p53-positive cholangiocytes fell from 61.1% to 7.5%; the proliferation index was 4.8-fold lower; and fibrosis staging dropped from a median of 4 to 2 ²⁰. The 2026 paper edited a different gene, the tetraspanin Ov-tsp-2, and reported in passing two facts that discipline the granulin story: Ov-GRN-1 is not detectable in the fluke's extracellular vesicles, and knocking out Ov-grn-1 is not lethal to the worm in vivo ⁸¹.

Four caveats belong on the page. The editing was mosaic, not a clean null: in the 2022 experiment the median indel rate in adult worms was 3.1% of sequenced alleles, ranging from 0% to 91%, while transcript levels fell much further — so no worm was a true knockout, and the explanation offered for the gap, that the plasmid reaches the multinucleated surface tissue where the gene is expressed, is an inference ^20,81. Second, worm burden: in a parallel 14-week infection, knockout hamsters carried 38.9% fewer worms, so less disease could partly reflect fewer parasites. Worms could not be counted in the 24-week cancer experiment itself, because those livers were fixed whole, so the confounder was never measured in the run that produced the cancer result; the authors measured fibrosis immediately around individual flukes to compare per worm instead (12.66% against 14.55%), which mitigates it without removing it ²⁰. Third, the intended comparator failed: the Ov-tsp-2 knockout turned out to be near-lethal, so that arm did not provide the clean control it was designed to be ²⁰. Fourth, the cancer counts themselves are small — 10 of 12 against 7 of 13 for any cancer, 8 of 12 against 6 of 13 for high-grade disease. The paper reports no significance test for them; a standard exact test on the published counts gives p ≈ 0.20 and p ≈ 0.43 (our calculation). The statistically clear results are the fibrosis, proliferation and TP53 measures, not the tumor count ²⁰.

The earlier work in this line survives the correction and matters independently. RNA interference against Ov-grn-1, done before CRISPR had been applied to liver flukes, suppressed expression by 95% at day three and retarded hyper-proliferation of cholangiocytes by 25% and of a cholangiocarcinoma line by 92%; the silenced flukes were also less viable in culture, which the authors read as the protein being needed for the adult worm's own survival in vitro ³². Laboratory-made Ov-GRN-1 is putatively taken into about 75% of cultured human cholangiocytes of the H69 line, speeds wound closure in mouse skin and promotes new blood vessels — the same properties that make it a wound-healing drug lead for the group that discovered it, which is a translational interest worth naming ³³. All of the knockout work so far comes from one collaborating group; we found no report of an independent replication, and the authors themselves describe heritable transgenic lines of the fluke as unlikely in the near future ²⁰.

Why fluke-associated bile duct cancer is a different cancer

If the fluke merely raised the odds of an ordinary cancer, the tumors would look the same as everyone else's. They do not. The first exome sequencing of fluke-associated tumors, eight read in full with 46 more screened for the genes that came out of them, found mutations in TP53, the gene for the p53 tumor suppressor protein, in 44.4% of cases, in KRAS and SMAD4, two genes that carry growth signals inside a cell, in 16.7% each, and in ten genes not previously implicated in this cancer ⁵⁸. A year later, exome sequencing of 209 cholangiocarcinomas, 108 of them fluke-related and 101 from other causes, found statistically significant differences in which genes were mutated: BAP1, IDH1 and IDH2 more often in the non-fluke cancers, and TP53 showing the reciprocal pattern ⁵⁹. The authors' conclusion was general and important: different causative exposures can induce distinct somatic alterations within the same tumor type ⁵⁹.

The larger follow-up made the split structural rather than incidental. Across 489 cholangiocarcinomas from 10 countries, with 71 whole genomes plus targeted sequencing, copy-number, expression and methylation data, integrative clustering produced four clusters that tracked cause more closely than anatomy. Fluke-positive cancers were enriched in clusters 1 and 2, which carried ERBB2 amplifications — ERBB2, also called HER2, is a growth-factor receptor — and TP53 mutations; fluke-negative cancers were enriched in clusters 3 and 4, which carried either many copy-number changes or else mutations in IDH1 and IDH2, metabolic enzyme genes, and BAP1, a tumor suppressor, together with rearrangements involving the growth-factor receptor gene FGFR2. Clusters 1 and 4 also differed in where the cell had put its chemical tags on the DNA, on CpG islands in one and on the shores beside them in the other ⁸². The phrase the authors used — that molecular landscapes differ radically by cause — is the cleanest available statement of why this parasite is not simply one more risk factor on a list.

Current work is filling in the texture and the trouble. Multi-region sequencing of 52 tumor samples from 13 patients with fluke-associated cancer found heavy intratumoral heterogeneity: between 48% and 90% of mutations were concentrated in each individual region of the tumor, with an average of 95 non-synonymous mutations per region, and driver mutations in TP53 and SMAD4 recurring ⁸³. In a panel of cell lines grown from fluke-positive patients, chromosome segregation errors were raised against a normal cholangiocyte line, though at lower rates than in colorectal or ovarian cancer; separately, low-pass whole-genome sequencing of 33 tumors from northeastern Thailand with matched normal tissue found whole-chromosome and subchromosomal alterations, and eight of those tumors carried focal amplifications or deletions of known cancer genes, including CCNE1, a cell-cycle gene, and one rare amplification of BRCA1, a DNA-repair gene ⁸⁴. A European series for contrast found extra copies of PBX1, a gene whose protein switches other genes on and off, in 20% of tumors, which the authors note tends to occur in European rather than fluke-associated Asian cancers ⁸⁵.

The practical reading is cautious. These are molecular differences between groups of tumors, established in sequencing studies, and they reinforce the causal argument by showing that the exposure leaves a mark. They do not yet mean a different treatment: targetable alterations such as FGFR2 fusions, IDH1 mutations and ERBB2 amplification are distributed unevenly across causes and anatomical subtypes, and the current review calls for large etiology-stratified genomic studies that do not yet exist ⁸⁶; we found no trial that has assigned therapy by fluke status. For a listener, the useful sentence is that a bile duct cancer from the Mekong basin and one from Europe are, at the level of their DNA, often different diseases wearing the same name. And the alterations that today's approved targeted drugs are aimed at — FGFR2 fusions, for which pemigatinib is approved in Korea, and IDH1 mutations, for which ivosidenib is — sit mainly in the fluke-negative groups, and the frequency of FGFR2 alterations itself varies by cause, so fewer fluke-associated patients are candidates for them ^82,87,88.

How it is found, and how it is missed

The standard test is to find eggs in feces, and the standard test is now the weak link. Light infection has become the norm — 99.1% of fluke-positive samples in Thailand's 2019 national survey were classed as low intensity, with a mean of 120.9 eggs per gram ⁶⁷ — and sensitivity falls with egg count. In a head-to-head comparison of routine methods, the formalin-ethyl acetate concentration technique found 91.0% of infections, including 94.3% of those under 50 eggs per gram, and outperformed Kato-Katz, a commercial concentrator kit and a direct smear; substituting the kit for either reference method was explicitly not recommended for light infections ⁸⁹. A modified one-step version of the concentration technique raised sensitivity to 95.6% against 87.9% for the conventional method, 55.5% for Kato-Katz, 48.3% and 35.1% for two commercial kits and 42.3% for a simple smear, and lowered the detection limit from 34 to 18 eggs per gram; in the field, 67% of samples below 50 eggs per gram were missed by the conventional technique ⁷⁴.

Sampling multiplies the problem. Against species-specific PCR as the reference in a Lao community, one Kato-Katz thick smear found 62.3% of infections and the concentration technique 49.3%, while three smears found 91.4% ⁷³. That result reverses the ranking the Thai comparisons gave, and a 2026 meta-analysis of 40 studies found no overall difference between the concentration technique and a single Kato-Katz smear for this fluke, odds ratio 1.09 (95% CI 0.61–1.92), with very high variation between studies ⁹⁰. What the methods agree on is that one sample, read once, misses many light infections: three samples is better practice and harder to run in a village. And the eggs themselves are ambiguous: mixed infection with minute intestinal flukes, particularly Haplorchis taichui, is common, the eggs look alike under a light microscope, and the two PCR assays developed to separate them found community prevalences of 63.9% for the liver fluke and 30.6% for the intestinal fluke in the same population ⁷³. In 2024 the size of that problem was measured directly: a DNA test using CRISPR as its detector confirmed the liver fluke in only 18 of 29 stool samples in which microscopy had reported its eggs, and the other 11 were most likely minute intestinal flukes ⁹¹. An egg count attributed to the liver fluke can therefore be partly someone else's.

Three newer approaches change the picture, and one of them changes the prevalence figures outright. A monoclonal antibody ELISA detects fluke excretory-secretory antigen in urine: applied alongside stool examination across 23 sub-districts in 21 Thai provinces, it gave 50.3% prevalence against 12.2%, with sensitivity 91.6% against 21.9% measured against the combined reference, and the gap was widest in the provinces with the lowest egg counts ¹³. That reference counted anyone positive on either test as infected, so it cannot say how often the antigen assay is positive in someone who is not infected, and it flatters the assay's sensitivity ¹³. Two independent checks support most of the extra positives all the same: reading five drops of fecal sediment instead of the usual two found eggs in 30.5% of people who were antigen-positive and egg-negative ⁶², and a rapid urine strip built on the same antibody, read like a pregnancy test, matched stool concentration at 94.2% sensitivity and 93.2% specificity with 2% cross-reaction with other worms, gave 53.2% prevalence against 54.0% for the laboratory assay in the field, and showed cure in 86.6% of people a month after praziquantel ⁶³. By 2025 those strips were being photographed and graded by a phone-based system that had processed more than 100,000 samples in the northeast, though its own authors say it has not been validated anywhere else ⁹². Species-specific PCR on stool found 19.9% where microscopy found 6.1% in a 2025 community study, and identified male sex as a risk factor with an adjusted odds ratio of 5.55 ⁶⁸. Antibody tests, by strip or ELISA, were positive in 26.1% and 31.6% of 307 rural participants where stool concentration found 16% ⁶⁹ — though detecting antibody cannot by itself separate current from past infection, which is why the assays that detect the worm's own products are preferred when the question is whether a worm is there now ^1,13. One practical warning: the point-of-care circulating cathodic antigen strip used for schistosomiasis cross-reacts with this fluke, so in areas where both occur a positive schistosomiasis strip may be reporting the wrong parasite ⁹³.

Finding the cancer is a separate problem from finding the worm, and it is solved differently — by looking for the scar rather than the parasite. Abdominal ultrasound grades periductal fibrosis by which ducts are involved, and the Cholangiocarcinoma Screening and Care Program (CASCAP) in northeast Thailand was built on it: a prospective cohort intended to enroll at least 150,000 high-risk people and to scan each of them at least once a year, with 85,927 enrolled and 55.0% scanned at the time of its first report, alongside a patient cohort ⁹⁴. As later redesigned, the program screens for the worm with the urine antigen test and for the damage it leaves with ultrasound ⁹⁵. In 751,061 screened participants, 15.35% had fibrosis of some grade and 0.11% had cancer; fibrosis of the segmental ducts carried an adjusted odds ratio for cancer of 1.4 (95% CI 1.03–1.91), of the main duct 2.52 (1.38–4.58), and of the peripheral ducts alone none at all, 0.94 (0.74–1.20) ⁷². Earlier ultrasound work in 3,359 egg-positive people had put advanced periductal fibrosis at 23.6%, rising with infection intensity ²².

• Stool microscopy for eggs (concentration technique, Kato-Katz smear, simple smear)Eggs are concentrated from feces and counted under a microscope, and the count is reported as eggs per gram of stool. The formalin-ethyl acetate concentration technique, or FECT, was the most sensitive routine method in two Thai comparisons ^74,89, but not in a Lao study scored against DNA testing ⁷³, and a 2026 meta-analysis of 40 studies found no overall difference between concentration and a single Kato-Katz smear for this fluke ⁹⁰. How carefully it is done matters as much as which method: reading five drops of sediment instead of the usual two raised sensitivity from 58.2% to 67% ⁶².The reference method for surveys and for individual diagnosis in endemic countries ¹FECT found 91.0% of infections overall in one Thai comparison, including 94.3% of those under 50 eggs per gram ⁸⁹. A modified one-step FECT reached 95.6% against 87.9% for the conventional version, 55.5% for Kato-Katz and 42.3% for a simple smear, and detected down to 18 eggs per gram against 34 ⁷⁴.Light infections, which are now most infections: in Thailand's 2019 national survey 99.1% of fluke-positive samples were low intensity, mean 120.9 eggs per gram ⁶⁷. In the one-step FECT study, 67% of samples under 50 eggs per gram were missed by the conventional technique ⁷⁴. One Kato-Katz smear found 62.3% of infections against PCR and the concentration technique 49.3%, while three smears found 91.4% ⁷³.
• Telling the eggs apart from minute intestinal flukesSeveral small intestinal flukes, above all Haplorchis taichui, lay eggs that look much like this one down a microscope: 29.03 ± 1.48 by 14.94 ± 0.91 µm against 26.34 ± 1.65 by 15.54 ± 0.69 µm, with widths that do not differ significantly ¹¹⁷. Size and shape vary too much within each species to separate them, and only surface detail, seen after special staining or under an electron microscope, reliably does ^117,118. Mixed infections are common ⁷³.A problem rather than a test; it is why species-specific PCR exists ⁷³Species-specific PCR in a Lao community found 93.7% (95% CI 85.8–97.9) of O. viverrini infections and 73.3% of H. taichui, and put community prevalence at 63.9% and 30.6% respectively ⁷³. In 2024, a DNA test using CRISPR as its detector confirmed the liver fluke in only 18 of 29 stool samples in which microscopy had reported its eggs; the other 11 were most likely minute intestinal flukes ⁹¹.Microscopy alone cannot separate them reliably, so an egg count attributed to the liver fluke may belong partly to another species ⁷³.
• Urine antigen assay (monoclonal antibody ELISA)This looks for fluke excretory-secretory protein excreted in urine, so a positive means a living worm rather than a past exposure, and the sample is easy to collect ¹³.Mass screening and research, now including a rapid urine strip used in the field ^13,63,92Across 23 sub-districts in 21 Thai provinces, 19,465 urine and 18,929 stool samples gave 50.3% prevalence by urine antigen against 12.2% by stool — a fourfold difference — with sensitivity 91.6% against 21.9% for stool ¹³. That comparison counted anyone positive on either test as infected, a design that cannot measure how often the antigen assay calls an uninfected person positive and that flatters its sensitivity ¹³; the strip version of the same assay, scored against stool concentration, gave 94.2% sensitivity and 93.2% specificity, with 2% cross-reaction with other worms ⁶³.The gap is largest exactly where infection is light, so the assay reclassifies low-endemicity areas rather than high ones; antigen-based site prevalence ranged from 22.2% to 71.4%, and egg counts and antigen concentrations correlated with each other across sites ¹³.
• Antibody tests and DNA amplificationAntibody tests — an enzyme-linked immunosorbent assay, or ELISA, which detects a target molecule with an antibody that produces a color change, and an immunochromatographic strip, which shows a line like a home pregnancy test — detect the immune response, which can outlast the worm ⁶⁹. PCR, the polymerase chain reaction, copies a chosen stretch of DNA until there is enough to read; here the stretch is ITS2, a spacer region that differs between fluke species ⁶⁸.Screening and research, and for light infections that microscopy misses ^68,69In 307 rural Thai participants, stool concentration found eggs in 16%, a strip test was positive in 26.1% and ELISA in 31.6% ⁶⁹. In 246 people in Sakon Nakhon in 2025, microscopy found 6.1% — all very light, under 500 eggs per gram — while ITS2 PCR found 19.9%, more than three times as many ⁶⁸.Antibody positivity does not prove a living worm, and DNA can persist; antibody tests are used for screening and PCR mainly in reference laboratories ^1,68. A point-of-care antigen strip designed for schistosomiasis also cross-reacts with this fluke, which matters where both occur ⁹³.
• Abdominal ultrasound for periductal fibrosisUltrasound looks not for the worm but for the scarring it leaves around the bile ducts, graded by which ducts are involved; this is the marker the Thai screening program is built on ^72,94.Population screening of people at risk of bile duct cancer, in the Cholangiocarcinoma Screening and Care Program, or CASCAP ^94,95In 751,061 screened participants, 15.35% had periductal fibrosis and 0.11% had cancer; fibrosis of the segmental ducts carried an adjusted odds ratio of 1.4 (95% CI 1.03–1.91) and of the main duct 2.52 (1.38–4.58) ⁷². Cancers found by screening had 53.9% five-year survival and a median survival of 67.2 months, against 21.9% and 15.6 months for people who came in with symptoms ¹⁰⁴.Fibrosis of the smallest peripheral ducts was not associated with cancer at all (adjusted odds ratio 0.94, 95% CI 0.74–1.20), so the mildest grade carries little information ⁷², and the comparison of screened against symptomatic patients is not a randomized trial, so part of the survival difference is the earlier clock start.

What is done about it, and what that achieves

Praziquantel kills the worm reliably. A network meta-analysis of 26 trials from 25 studies and 3,340 participants put the highest predicted cure rate for O. viverrini at 93.8% (95% CI 85.7–97.5) for 50 mg/kg followed by 25 mg/kg in one day, 92.1% (64.9–98.6) for a single 50 mg/kg dose, and 89.8% (77.5–95.8) for single-dose tribendimidine 400 mg — with the caveat, stated by the authors, that most included studies were of low quality, especially those published before 2000 ⁵. Tribendimidine, registered in China for soil-transmitted worms, is the main alternative in development ¹. Repeated dosing may not be entirely free. In 490,969 screened people, those reporting more than three courses of praziquantel had 1.25 times the odds of a kidney abnormality on ultrasound (95% CI 1.02–1.52) — in absolute terms, 1.5% against 1.2% overall, rising stepwise from 1.2% with one course ⁹⁶. Treatment history was self-reported, and liver fluke infection is itself associated with kidney injury in people and animals ⁹⁷, so repeated courses may partly stand for repeated infection; the authors judge the drug's benefit for an infected person to outweigh it ⁹⁶.

Killing the worm is not the same as undoing the damage, and this is where the evidence gets uncomfortable. In hamsters, praziquantel given four weeks after infection significantly reduced bile duct lesions, but at twelve and twenty weeks it did not; the authors concluded that the fluke's enhancement of bile duct carcinogenesis is “rapid and to a large degree irreversible” ⁹⁸. In people, urinary 8-oxodG — the oxidative DNA damage marker — fell significantly after treatment in infected participants without periductal fibrosis but was unchanged in the 52 who had it; of those with fibrosis, 65.3% became fibrosis-negative on ultrasound a year later while 34.6% had it at both time points ⁷⁶. Treatment is worth giving. It is not an eraser.

And reinfection resets the clock. In a cohort of 612 people in northeast Thailand monitored for 24 weeks after selective treatment, baseline prevalence was 41.0% by urine antigen against 8.1% by stool; the incidence of infection was 64.6 per 100 person-years and the reinfection rate after praziquantel 63.7 per 100 person-years by antigen, against 5.9 by stool — a tenfold difference depending on the test ⁹⁹. Raw fish consumption in the previous six months carried an adjusted risk ratio of 7.52, and previous praziquantel treatment 1.49 ⁹⁹. A separate five-year follow-up of 418 people found 10.8% reinfected ¹⁰⁰. Both numbers are real; they measure different things over different intervals with different tests, and neither licenses a single headline percentage.

Two things appear to work, and a third has been tested with mixed results. First, community control: the Lawa Lake program in Khon Kaen province combined anthelmintic treatment, intensive health education in villages and schools, ecosystem monitoring and active community participation over fifteen years, and by its architects' 2025 account brought infection around the lake from about 60% to under 5%, with infection in the Bithynia snails and the cyprinid fish now undetectable; the model is being scaled up as a training hub in Thailand and the wider Mekong region ^101,102. The earlier report of the same program recorded infection falling to about a third of the 50% baseline average ¹⁰³.

Second, screening: in 711 histologically proven cases, those found by ultrasound screening had 53.9% five-year survival and a median survival of 67.2 months against 21.9% and 15.6 months for people presenting with symptoms ¹⁰⁴, and in a single-center surgical series of 1,091 patients, early-stage disease rose from 16.0% to 29.1% and median survival from 14 to 40 months between 2002–2013 and 2014–2021 — a span over which surgical strategy also changed substantially. Only 66 patients, 15.2% of the later period, had been referred by screening at all, and within that period survival did not differ significantly between screened and usual presentation (51 against 38 months, p = 0.06) ¹⁰⁵. Third, school education, which changed what children knew but not what they did: a cluster-randomized trial of a cartoon-based school program across Cambodia, Laos and Thailand raised knowledge and attitude scores by about 19 and 25 percentage points in Cambodia and 19 and 14 in Laos — but scores fell by 23.3 and 15.8 points in Thailand, and behavior scores improved in no country ¹⁰⁶.

Two cautions sit over all of it. The Lawa evaluation was written by the architects of the program, with no independent assessment yet ^101,102; the first randomized test is under way — a cluster-randomized trial in 18 communities across Cambodia, Laos and Thailand, comparing one round of praziquantel and albendazole against the same treatment plus the Lawa model and the Magic Glasses school package, with the incidence of infection at 12 months as its endpoint — but only its protocol has been published, in 2024, its results are not yet in the literature, and its authors include the program's founder ^106,107. And the comparison of screened against symptomatic patients is not a randomized trial either, so part of the survival advantage is the earlier start of the clock rather than a life extended ¹⁰⁴. The deeper caution is the one a transdisciplinary field study put plainly: treating liver fluke infection and bile duct cancer as the same issue is inadvisable, because the two are shaped by overlapping but different factors, and an intervention aimed only at the parasite may not move the cancer as much as expected ¹⁰⁸.

What was believed, and is not

This organism has the most completely evidenced case of any parasite for causing a human cancer. Its pooled odds ratio is in fact a little lower than Clonorchis sinensis's, 3.69 against 4.49 ¹⁰, but no other parasite has epidemiology, an animal model and a gene-editing test standing behind it together. And it is still described wrongly in at least seven places, sometimes by overstating the link and sometimes by understating it. The corrections below do not weaken the classification. They tighten what can honestly be said: that this is a Group 1 human carcinogen whose effect runs through decades of inflammation in company with diet, bacteria, tobacco and alcohol, in a population where the cancer rate is now falling and the infection rate is probably higher than the official figures say.

Each card below names the belief, the correction, and the study that made the correction. Where a widely repeated number has no measurement behind it, that is said as well, because a figure quoted often enough starts to sound like data. Two of these corrections cut against the parasite's reputation and two cut for it, which is the point: the same literature that disciplines the folklore also keeps the organism firmly in Group 1, and a reader who takes away only one direction will be wrong. The last card is not a belief at all but a retraction notice, kept here so that its numbers are not quoted again by accident.

What is still unknown

The decisive trial does not exist and probably never will. No randomized comparison of mass praziquantel against no treatment, with cholangiocarcinoma incidence at fifteen or twenty years as the endpoint, was found in the literature reviewed for this page — and that is the only design that would settle whether clearing the fluke lowers the cancer rate. The trials that do exist measured cure by egg clearance ⁵. What exists instead is a set of observational findings pointing in opposite directions — treatment associated with more cancer in a hospital case-control study ^14,112, with less fibrosis in a screening cohort ¹⁵, and with no fall in oxidative DNA damage in people who already have fibrosis ⁷⁶ — together with a 2013 meta-analysis that found no significant association with treatment ⁶⁰ and a 2016 hamster study in which repeated infection followed by praziquantel produced no cancer ⁶¹, against an older animal result suggesting the enhancement is largely irreversible once established ⁹⁸. The page should stage that conflict rather than resolve it.

The nitrosamine chain is asserted at one end and measured at the other, with the join missing. Nitrosamines have been measured in Thai fermented and dried fish sampled from four major regions of the country, including N-nitrosodimethylamine up to 66.5 µg per kilogram in salted and dried fish ⁴¹, and infected people show higher endogenous nitrosation after a proline load — though in that same study, cancer incidence across five areas did not track any of the nitrosation measures ⁴⁰. The hamster needs the chemical to get cancer ^38,52. But no study has measured an individual's dietary nitrosamine intake and followed that individual to a cancer; the step from “hamsters were given dimethylnitrosamine” to “villagers get nitrosamines from their fish” remains inference supported by food chemistry, not a measured exposure-outcome link.

The mechanistic work has three open edges. The mechanical strand has never been isolated from inflammation in a quantitative measurement; the best available proxy uses surgical ligation and no parasite ²⁸. The granulin knockout comes from a single collaborating group and has not, so far as our search goes, been reported by an independent laboratory; the editing is mosaic rather than a true null, knockout hamsters carried 38.9% fewer worms in a parallel 14-week infection, and the authors themselves describe heritable transgenic lines of the fluke as unlikely in the near future ²⁰. And nobody has measured Ov-GRN-1 concentrations in human bile against subsequent cancer risk, so the molecule's role in people rests on animal and cell-culture work plus the detection of fluke vesicle proteins in human bile ³⁶.

Four smaller questions matter for the script. Reinfection has two defensible numbers — 63.7 per 100 person-years by urine antigen over 24 weeks ⁹⁹ and 10.8% at five years by stool ¹⁰⁰ — and no single figure should be quoted as the reinfection rate. Whether established periductal fibrosis regresses after cure is partly answered (65.3% became ultrasound-negative, 34.6% did not) and the determinants of that split are unknown ⁷⁶. Whether the molecular distinctness of fluke-associated cancer should change treatment is untested ⁸⁶. And it is not established what share of the falling Khon Kaen cancer rate reflects genuinely less transmission rather than registry factors such as coding and case ascertainment, which the registry analysis does not address ¹²; nor is it settled whether the fall holds beyond Khon Kaen, since a ten-year screening-program analysis across 20 provinces describes the cancer as still rising ¹¹¹. The infection prevalence that control programs report is a separate matter, and it is understated by stool testing ¹³.

How it fits the season's question

The season asks whether the cancers attributed to parasites are the real number or only the number someone has looked for. This organism is the most completely evidenced case the affirmative has, and it answers the question in both directions at once. The classification is sound: a working group that sat for eight days in Lyon in 1994 placed it in Group 1, the 2009 review reaffirmed it, and IARC still counts it among its 13 Group 1 infectious agents in 2026 ^7,8,9,18. The biological case is unusually complete for a parasite: a dose-response measured in a population of 12,311 people ⁵⁴, an odds ratio of 5.0 in a case-control study that found no association with hepatitis B, with aflatoxin or with any particular dietary pattern ⁵³, a pooled odds ratio of 3.69 for this species from a meta-analysis that pooled 22 studies across all three liver flukes ¹⁰, an animal model in which the fluke plus a chemical produces the tumor and neither does alone ^38,52, and a single parasite gene whose partial disruption sharply reduced the precancerous damage in hamsters, with a trend toward less cancer ²⁰.

And yet the number attached to it has always depended on the test used to find the parasite. The most important finding on this page for the season's question is not about cancer at all: when a urine antigen assay was run beside stool microscopy in 21 Thai provinces, prevalence went from 12.2% to 50.3%, and the authors concluded that previous surveys may have underestimated how much opisthorchiasis there is ¹³ — an upward revision supported, though not proven, by independent checks with the older method and with a rapid strip ^62,63. Light infections — now 99% of them in Thailand ⁶⁷ — are exactly what the old test misses ⁷⁴. Here, uniquely in the season, the looking has been improved, and the number went up.

The honest reading of the cancer side pulls the other way, which is what makes this entry useful rather than merely alarming. The biggest measured risk factor for cholangiocarcinoma in the lower Mekong, in a meta-analysis of 18 studies, was not the fluke: it was the combination of alcohol and smoking, at an odds ratio of 11.1 (95% CI 5.63–21.92) — which the authors describe as a greater risk factor than exposure to the parasite ⁵⁶. Two cautions belong beside that comparison. Exposure to the fluke in studies of this kind is established by stool eggs or by antibodies, and both tests call many genuinely infected people uninfected ^13,73; misclassification in that direction pulls an odds ratio toward 1, so the fluke's figure is more likely an underestimate than an overestimate. And the same 2024 meta-analysis that produced the odds ratio also reports a pooled relative risk of 10.43 (95% CI 2.90–37.47) for the three flukes together ¹⁰. Diabetes adds to the fluke's effect, with an adjusted odds ratio of 2.36 for people with both against neither ⁷⁵. The fluke carries bacteria that worsen the pathology ⁴². And a transdisciplinary field program in five northeastern Thai villages argued that treating liver fluke infection and bile duct cancer as one issue is inadvisable ¹⁰⁸. The parasite is a cause; it is not the whole cause, and a program aimed only at it will not capture the whole cancer.

What this page can say plainly, in the house voice: Opisthorchis viverrini causes cancer in people, the evidence for that is as complete as it gets for any parasite, and the strongest single piece of it is an experiment in which partly disrupting one of the worm's own genes reduced fibrosis, cell proliferation and p53 mutation in the animals it infected, with fewer and less severe cancers in groups too small to settle that last point ²⁰. The attributed burden is nonetheless soft at both ends — a prevalence that quadruples when the test improves ¹³, and a cancer incidence that is falling along a birth-cohort pattern for reasons the registry analysis does not establish ¹² — and IARC's own attributable count, 3,800 intrahepatic cancers a year worldwide for both Asian liver flukes together, is a floor set by which tumors get counted ¹⁸. A general framework published in 2026 puts the fluke in its proper place: most worms that cause chronic inflammation never cause cancer, and this one does because mechanical injury, secreted mitogens, dietary nitrosamines, disturbed bacterial communities in the bile ducts — of which the Helicobacter the fluke carries is the best-studied example ⁴² — and failed repair converge in the same duct for decades ^27,64. That is a conditional, specific claim, and it is the kind the show should make.

Where it connects

Sources

118 sources, numbered as they are cited. Every one was checked against PubMed or its publisher before it was cited here; the note under each says what it shows and what it does not.

1. 1
   Qian MB, Keiser J, Utzinger J, Zhou XN. Clonorchiasis and opisthorchiasis: epidemiology, transmission, clinical features, morbidity, diagnosis, treatment, and control.Clin Microbiol Rev · 2024 · 37(1):e0000923doi:10.1128/cmr.00009-23 · PMID 38169283

   The current reference review of all three liver flukes: the shared life cycle, the regional split, and chronic infection associated with biliary complications, most importantly cholangiocarcinoma. It states association rather than causation for that link, and names tribendimidine as a drug warranting further development.

2. 2
   Buisson Y. [Control of Opisthorchis viverrini infection for cholangiocarcinoma prevention].Bull Soc Pathol Exot · 2017 · 110(1):61-67doi:10.1007/s13149-017-0544-8 · PMID 28105582

   States the IARC chronology directly (O. viverrini in 1994, C. sinensis in 2009), the life cycle with Bithynia and cyprinid hosts, the 10-to-30-year adult lifespan, the three intertwined mechanisms with nitrosamines as cofactor, and five-year survival under 5%. A review, in French, with no new data.

3. 3
   Banales JM, Cardinale V, Carpino G, et al. Expert consensus document: Cholangiocarcinoma: current knowledge and future perspectives consensus statement from the European Network for the Study of Cholangiocarcinoma (ENS-CCA).Nat Rev Gastroenterol Hepatol · 2016 · 13(5):261-80doi:10.1038/nrgastro.2016.51 · PMID 27095655

   The European consensus definition used here for the disease itself: a heterogeneous group of malignancies with features of biliary tract differentiation, the second most common primary liver tumor, with rising incidence and high mortality from aggressiveness, late diagnosis and refractory behavior. A consensus statement, not primary data.

4. 4
   Dao HTT, Dermauw V, Gabriël S, et al. Opisthorchis viverrini infection in the snail and fish intermediate hosts in Central Vietnam.Acta Trop · 2017 · 170:120-125doi:10.1016/j.actatropica.2017.02.028 · PMID 28242064

   Direct field measurement of both intermediate hosts: of 12,000 snails only the two Bithynia species shed cercariae, at 0.86% and 0.14%, while metacercariae were found in 10 of 12 fish species, with Carassius auratus at 74.0%. One province in one year; infection in snails tracked habitat rather than species.

5. 5
   Qian MB, Patel C, Palmeirim MS, et al. Efficacy of drugs against clonorchiasis and opisthorchiasis: a systematic review and network meta-analysis.Lancet Microbe · 2022 · 3(8):e616-e624doi:10.1016/S2666-5247(22)00026-X · PMID 35697047

   Gives the figure of about 25 million people infected by the three species and the predicted cure rates for each praziquantel and tribendimidine regimen. The authors state that most included studies were of low quality, especially those before 2000; cure is measured by egg clearance, not by cancer outcome.

6. 6
   Sripa B, Suwannatrai AT, Sayasone S, Do DT, Khieu V, Yang Y. Current status of human liver fluke infections in the Greater Mekong Subregion.Acta Trop · 2021 · 224:106133doi:10.1016/j.actatropica.2021.106133 · PMID 34509453

   The current geographic picture: more than 12 million infected with O. viverrini, 3.5 to 5 million with C. sinensis, Thailand's prevalence falling from over 15% in 1996 to 2.2% in 2019, up to 47.5% recorded in Cambodia, and the high-risk area shifting away from northeastern Thailand. A review of surveys of varying method and quality.

7. 7
   IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Schistosomes, liver flukes and Helicobacter pylori.IARC Monogr Eval Carcinog Risks Hum · 1994 · 61:1-241PMID 7715068

   The bibliographic record for the 1994 monograph that evaluated the liver flukes: it confirms the volume number, the 241-page extent, the working group meeting in Lyon from 7 to 14 June 1994, and the organisms in scope. The record carries no abstract, so the group assignments themselves cannot be quoted from it.

8. 8
   Bouvard V, Baan R, Straif K, et al. A review of human carcinogens--Part B: biological agents.Lancet Oncol · 2009 · 10(4):321-2doi:10.1016/s1470-2045(09)70096-8 · PMID 19350698

   The contemporaneous announcement of the 2009 biological-agents working group, by IARC Monographs staff. PubMed types it as news and it carries no abstract, so it is cited here only for the date and authorship of that review, not for any finding.

9. 9
   IARC. Biological agents. Volume 100 B. A review of human carcinogens.IARC Monogr Eval Carcinog Risks Hum · 2012 · 100(Pt B):1-441PMID 23189750

   The record for the 441-page volume 100B, published 2012, which re-reviewed agents already in Group 1 and covers opisthorchiasis, clonorchiasis and urogenital schistosomiasis. No abstract, so group assignments are not quotable from this record.

10. 10
    Huang YL, Zhang KY, Sun YL, Qian MB, Wang Z. The risk of hepatobiliary complications in Clonorchis and Opisthorchis infection: A systematic review and meta-analysis.Acta Trop · 2024 · 260:107457doi:10.1016/j.actatropica.2024.107457 · PMID 39521195

    The largest pooled human estimate: 22 studies, 34,367 participants, overall cholangiocarcinoma odds ratio 4.24 and 3.69 for O. viverrini specifically. These are pooled observational associations, not a demonstration of causation, and no separate estimate is given for O. felineus.

11. 11
    Kamsa-ard S, Wiangnon S, Suwanrungruang K, et al. Trends in liver cancer incidence between 1985 and 2009, Khon Kaen, Thailand: cholangiocarcinoma.Asian Pac J Cancer Prev · 2011 · 12(9):2209-13PMID 22296358

    The registry numbers: of 18,589 liver cancers, 58% were cholangiocarcinoma, with an age-standardized rate of 44.3 per 100,000 in men (95% CI 38.9-49.7) and 17.6 in women, and annual values from 16.8 to 62.0 in men. Morphological verification by cytology or histology was only 10.8%, which is the main weakness of these figures.

12. 12
    Kamsa-Ard S, Santong C, Kamsa-Ard S, et al. Decreasing trends in cholangiocarcinoma incidence and relative survival in Khon Kaen, Thailand: An updated, inclusive, population-based cancer registry analysis for 1989-2018.PLoS One · 2021 · 16(2):e0246490doi:10.1371/journal.pone.0246490 · PMID 33592053

    The reversal, by three independent statistical methods: annual percent change -3.1% in men and -2.4% in women, a birth-cohort effect with an incidence rate ratio of 0.09 for men born in 1998 against 1966, a 2028 projection of 7.6 per 100,000 in men, and five-year relative survival of 10.9%. It does not establish how much of the fall is better control rather than better or worse counting.

13. 13
    Kopolrat KY, Worasith C, Wongphutorn P, et al. Large-scale epidemiology of opisthorchiasis in 21 provinces in Thailand based on diagnosis by fecal egg examination and urine antigen assay and analysis of risk factors for infection.PLoS Negl Trop Dis · 2025 · 19(7):e0013095doi:10.1371/journal.pntd.0013095 · PMID 40668872

    The study that revises the prevalence of this infection: 20,322 people enrolled across 23 sub-districts in 21 provinces, 50.3% positive by urine antigen against 12.2% by stool examination, sensitivity 91.6% against 21.9% on a combined reference, with the gap largest in low-endemicity areas. The authors conclude that previous surveys may have underestimated the extent of opisthorchiasis in Thailand. Because anyone positive on either test counted as infected, the design cannot measure the antigen assay’s specificity and flatters its sensitivity.

14. 14
    Kamsa-Ard S, Luvira V, Pugkhem A, et al. Association between praziquantel treatment and cholangiocarcinoma: a hospital-based matched case-control study.BMC Cancer · 2015 · 15:776doi:10.1186/s12885-015-1788-6 · PMID 26496745

    210 cases and 840 matched controls: a significant dose-response with odds ratios of 1.49, 1.82 and 2.30 for one, two and more than two courses of praziquantel, and 1.23 per additional course. The authors themselves suggest repeated treatment may be a surrogate marker for eating raw fish, so this is not evidence that the drug causes cancer.

15. 15
    Intajarurnsan S, Khuntikeo N, Chamadol N, Thinkhamrop B, Promthet S. Factors Associated with Periductal Fibrosis Diagnosed by Ultrasonography Screening among a High Risk Population for Cholangiocarcinoma in Northeast Thailand.Asian Pac J Cancer Prev · 2016 · 17(8):4131-6PMID 27644673

    The direct contradiction of the praziquantel-harm signal: in 55,246 screened people, periductal fibrosis prevalence was 33.0% and more than two praziquantel treatments was associated with 46% lower odds of it (adjusted odds ratio 0.54). The same analysis found diabetes apparently protective, which the authors flag as requiring further investigation and which signals confounding.

16. 16
    Young ND, Nagarajan N, Lin SJ, et al. The Opisthorchis viverrini genome provides insights into life in the bile duct.Nat Commun · 2014 · 5:4378doi:10.1038/ncomms5378 · PMID 25007141

    The draft genome, 634.5 Mb, with metabolic pathways adapted to a lipid-rich diet from bile and cholangiocytes, and additional evidence that the fluke secretes proteins that modulate host cell proliferation. A genomic resource; no clinical findings.

17. 17
    Laoprom N, Saijuntha W, Sithithaworn P, et al. Biological variation within Opisthorchis viverrini sensu lato in Thailand and Lao PDR.J Parasitol · 2009 · 95(6):1307-13doi:10.1645/GE-2116.1 · PMID 19658451

    Confirms biologically what molecular work had suggested: at least two cryptic species, six genetic groups tracking five river wetland systems, with significant differences between populations in worm recovery and fecundity. Hamster infection experiments; it does not test whether the groups differ in carcinogenicity.

18. 18
    Rumgay H, Georges D, Huang Y, et al. Global burden of cancer attributable to infections in 2024: a worldwide incidence analysis.Lancet Oncol · 2026 · 27(10):1237-1248doi:10.1016/S1470-2045(26)00307-4 · PMID 42805198

    IARC's own current burden analysis, and the best verification of which infectious agents are in Group 1 today. Its introduction states that the Monographs have classified 13 infectious agents as Group 1 — nine viruses, one bacterium and three parasites — and its own estimates cover 12 of them; hepatitis D virus, classified Group 1 in 2025, is not counted separately because the cancers it causes are a subset of those attributed to hepatitis B. It attributes 3,800 new intrahepatic cholangiocarcinomas worldwide in 2024 to O. viverrini and C. sinensis together (2% of that cancer) and 5,900 squamous cell carcinomas of the bladder to S. haematobium (33%), with no separate figure for each fluke and nothing counted for ducts outside the liver. The 13-agent count, the parasite case numbers and the registry-coverage figures are in the full text (PMC13619544), not the abstract; the abstract names only the 12 agents analyzed. Registry coverage behind the underlying incidence estimates is about 19% of the world's population and 2% of Africa's.

19. 19
    Arunsan P, Ittiprasert W, Smout MJ, et al. Programmed knockout mutation of liver fluke granulin attenuates virulence of infection-induced hepatobiliary morbidity.eLife · 2019 · 8:e41463doi:10.7554/eLife.41463 · PMID 30644359

    Published on the same day, in the same journal, as the first knockout in a blood fluke [91]; the paper’s own words are that together they pioneer programmed gene editing in parasitic flatworms. It edited the parasite's gene, not the host's. Edited flukes still matured in hamsters but caused less biliary thickening (145% of uninfected against about 500% for wild type at day 14) and 28% less collagen at day 60 (p ≤ 0.001). No carcinogen was given, so no cancer was measured; the paper used wild-type and mock-electroporated controls but no scrambled-guide control. The 145%, ~500% and 28% figures are in the Results of the full text (PMC6355195), not the abstract.

20. 20
    Chaiyadet S, Tangkawattana S, Smout MJ, et al. Knockout of liver fluke granulin, Ov-grn-1, impedes malignant transformation during chronic infection with Opisthorchis viverrini.PLoS Pathog · 2022 · 18(9):e1010839doi:10.1371/journal.ppat.1010839 · PMID 36137145

    The strongest causal evidence in the field, and the one experiment here with a cancer endpoint: cholangiocarcinoma in 10 of 12 controls against 7 of 13 knockout animals, mutant p53-positive cholangiocytes from 61.1% to 7.5%, proliferation 4.8-fold lower, fibrosis stage from a median of 4 to 2. Three limits: editing was mosaic (median 3.1% of adult alleles, range 0 to 91%), knockout hamsters carried 38.9% fewer worms, and the intended comparator gene turned out to be essential so its control arm failed. Every count here is from the full text (PMC9531791), not the abstract, which claims only that “fewer hamsters developed high-grade CCA.” The paper reports no significance test for cancer incidence: on the published counts a two-sided Fisher exact test gives p ≈ 0.20 for any cancer (10 of 12 against 7 of 13) and p ≈ 0.43 for high-grade disease (8 of 12 against 6 of 13) — our calculation. The 38.9% reduction in worm burden comes from the parallel 14-week experiment, not from the 24-week cancer run, in which the livers were fixed whole and worms could not be counted.

21. 21
    Wang YC, Law A, Namsanor J, Sithithaworn P. Examining ecosystem (dis-)services on liver fluke infection in rural Northeast Thailand.Infect Dis Poverty · 2023 · 12(1):38doi:10.1186/s40249-023-01079-y · PMID 37076937

    The cultural dimension measured: men ate koi pla and pla som on 64.5 and 40.4 days a year in two villages against 4.1 and 4.3 for women, consumption was driven mainly by cultural rather than nutritional factors, and taking part in dish-sharing cut the odds of willingness to stop (odds ratio 0.19). 120 questionnaires in two villages.

22. 22
    Mairiang E, Laha T, Bethony JM, et al. Ultrasonography assessment of hepatobiliary abnormalities in 3359 subjects with Opisthorchis viverrini infection in endemic areas of Thailand.Parasitol Int · 2012 · 61(1):208-11doi:10.1016/j.parint.2011.07.009 · PMID 21771664

    Of 8,936 people stool-examined and 3,359 egg-positive people scanned, 23.6% had advanced periductal fibrosis, more often men, with risk rising with infection intensity, and eleven suspected cancers were seen. Cross-sectional ultrasound findings in an infected population.

23. 23
    Sripa B, Thinkhamrop B, Mairiang E, et al. Elevated plasma IL-6 associates with increased risk of advanced fibrosis and cholangiocarcinoma in individuals infected by Opisthorchis viverrini.PLoS Negl Trop Dis · 2012 · 6(5):e1654doi:10.1371/journal.pntd.0001654 · PMID 22629477

    Plasma interleukin-6 was 58 times higher in people with advanced fibrosis and 221 times higher in people with bile duct cancer than in matched controls, with the highest quartile carrying 19 and 150 times the risk respectively. These are cross-sectional associations; high interleukin-6 tracked the pathology rather than the infection itself.

24. 24
    Chompo P, Sota P, Andityas M, Thongrin T, Sukon P, Tangkawattana S. Prevalence and zoonotic transmission of Opisthorchis viverrini in animal reservoir: a systematic review and meta-analysis in the greater Mekong Subregion.Vet Res Commun · 2026 · 50(5)doi:10.1007/s11259-026-11365-3 · PMID 42400729

    The measured size of the animal reservoir, and the reason treating only people leaves transmission running: pooled infection was 11.97% in cats against 1.89% in dogs, mean egg counts 151 against 46 per gram, and animal infection tracked human infection from place to place more closely for cats (R² 0.86) than dogs (R² 0.62). The overall pooled figure of 4.76% carries a prediction interval from 0.15% to 61.76%, which is another way of saying the studies disagree enormously.

25. 25
    Kaewkong W, Choochote W, Kanla P, et al. Chromosomes and karyotype analysis of a liver fluke, Opisthorchis viverrini, by scanning electron microscopy.Parasitol Int · 2012 · 61(3):504-7doi:10.1016/j.parint.2012.03.008 · PMID 22504416

    The first complete karyotype: six chromosomes, haploid n = 6 and diploid 2n = 12, measured from 0.84 to 2.84 µm. Cited here for the only precisely measured dimensions of this organism that this page could verify.

26. 26
    Wang YC, Grundy-Warr C, Namsanor J, et al. Masculinity and misinformation: Social dynamics of liver fluke infection risk in Thailand.Parasitol Int · 2021 · 84:102382doi:10.1016/j.parint.2021.102382 · PMID 33984515

    Four villages in Kalasin province: 76.9% gave embedded cultural practice as the reason for eating raw fish, about 30% said they would not stop regardless, 75.6% of men ate koi pla against 42.7% of women, and only 15.3% correctly identified raw fish as the source while 84.2% thought other raw foods transmitted it. Questionnaires and interviews, not infection measurements.

27. 27
    Smout MJ, Laha T, Chaiyadet S, Brindley PJ, Loukas A. Mechanistic insights into liver-fluke-induced bile-duct cancer.Trends Parasitol · 2024 · 40(12):1183-1196doi:10.1016/j.pt.2024.10.012 · PMID 39521672

    The current synthesis of mechanism: secreted growth factors, digestive enzymes and extracellular vesicles combined with aberrant inflammation and repeated wounding where the parasite grazes on the epithelium. Written by the investigators who generated most of the primary evidence it assesses, which is an interpretive rather than a financial conflict.

28. 28
    Thamavit W, Pairojkul C, Tiwawech D, Itoh M, Shirai T, Ito N. Promotion of cholangiocarcinogenesis in the hamster liver by bile duct ligation after dimethylnitrosamine initiation.Carcinogenesis · 1993 · 14(11):2415-7doi:10.1093/carcin/14.11.2415 · PMID 8242874

    The best available test of mechanical injury, using no parasite at all: nitrosamine initiation plus surgical bile duct ligation gave 60.9% cholangiofibrosis, 21.7% cystadenomas and 39.1% cholangiocarcinoma at week 40, against 5% cholangiofibrosis and no tumors with the chemical alone. It bounds the mechanism; it does not measure the fluke's own contribution.

29. 29
    Pinlaor S, Ma N, Hiraku Y, et al. Repeated infection with Opisthorchis viverrini induces accumulation of 8-nitroguanine and 8-oxo-7,8-dihydro-2'-deoxyguanine in the bile duct of hamsters via inducible nitric oxide synthase.Carcinogenesis · 2004 · 25(8):1535-42doi:10.1093/carcin/bgh157 · PMID 15059927

    The inflammation strand with named molecular markers: both DNA lesions rose in the duct lining in the order three infections above two above one, after inflammatory cells had subsided, with the lining itself expressing inducible nitric oxide synthase by day 90. Hamster data; no cancer endpoint.

30. 30
    Pinlaor S, Hiraku Y, Yongvanit P, et al. iNOS-dependent DNA damage via NF-kappaB expression in hamsters infected with Opisthorchis viverrini and its suppression by the antihelminthic drug praziquantel.Int J Cancer · 2006 · 119(5):1067-72doi:10.1002/ijc.21893 · PMID 16570287

    One week of praziquantel did not kill all parasites at days 14 and 30 yet nearly abolished both DNA lesions in the duct lining and lowered nuclear factor kappa B and the enzyme, which the authors read as an anti-inflammatory effect of the drug itself. Hamster data; the suggested cancer prevention is inferred, not measured.

31. 31
    Smout MJ, Laha T, Mulvenna J, et al. A granulin-like growth factor secreted by the carcinogenic liver fluke, Opisthorchis viverrini, promotes proliferation of host cells.PLoS Pathog · 2009 · 5(10):e1000611doi:10.1371/journal.ppat.1000611 · PMID 19816559

    The discovery paper for Ov-GRN-1: a homolog of human granulin in the fluke's secretions, expressed in gut and tegument, detected on biliary cells of infected hamsters, driving proliferation at nanomolar concentrations, with antibodies against it inhibiting most of the proliferative activity of the whole secretion mix. In vitro and animal localization; no cancer endpoint. The group also develops this molecule as a wound-healing drug lead.

32. 32
    Papatpremsiri A, Smout MJ, Loukas A, Brindley PJ, Sripa B, Laha T. Suppression of Ov-grn-1 encoding granulin of Opisthorchis viverrini inhibits proliferation of biliary epithelial cells.Exp Parasitol · 2015 · 148:17-23doi:10.1016/j.exppara.2014.11.004 · PMID 25450776

    The pre-CRISPR experiment: RNA interference suppressed the gene by 95% at day 3 and retarded hyper-proliferation by 25% in cholangiocytes and 92% in a cholangiocarcinoma line. It also claimed the protein was essential for adult survival in vitro, a claim the later gene-editing work does not support.

33. 33
    Smout MJ, Sotillo J, Laha T, et al. Carcinogenic Parasite Secretes Growth Factor That Accelerates Wound Healing and Potentially Promotes Neoplasia.PLoS Pathog · 2015 · 11(10):e1005209doi:10.1371/journal.ppat.1005209 · PMID 26485648

    What the molecule does to host cells: internalized by about 75% of cultured human cholangiocytes, inducing wound-healing and cancer-pathway gene expression, accelerating wound closure in mouse skin and angiogenic in an embryo assay. The same paper proposes it as a therapeutic wound-healing agent, which is a translational interest in its potency.

34. 34
    Suttiprapa S, Loukas A, Laha T, et al. Characterization of the antioxidant enzyme, thioredoxin peroxidase, from the carcinogenic human liver fluke, Opisthorchis viverrini.Mol Biochem Parasitol · 2008 · 160(2):116-22doi:10.1016/j.molbiopara.2008.04.010 · PMID 18538872

    A 212-amino-acid antioxidant enzyme cloned from the fluke, detected in bile fluid, in bile duct lining two weeks after infection, and in secondary bile ducts the flukes are too large to reach. It explains how the worm survives the radicals inflammation throws at it; it does not measure a cancer effect.

35. 35
    Matchimakul P, Rinaldi G, Suttiprapa S, et al. Apoptosis of cholangiocytes modulated by thioredoxin of carcinogenic liver fluke.Int J Biochem Cell Biol · 2015 · 65:72-80doi:10.1016/j.biocel.2015.05.014 · PMID 26007234

    Fluke thioredoxin entered cultured human cholangiocytes, blocked the apoptosis that hydrogen peroxide would otherwise cause, shifted apoptotic and anti-apoptotic gene expression, and bound apoptosis signal-regulating kinase 1. Cell culture only; the inference that blocking apoptosis facilitates carcinogenesis is the authors' framing.

36. 36
    Chaiyadet S, Sotillo J, Smout M, et al. Carcinogenic Liver Fluke Secretes Extracellular Vesicles That Promote Cholangiocytes to Adopt a Tumorigenic Phenotype.J Infect Dis · 2015 · 212(10):1636-45doi:10.1093/infdis/jiv291 · PMID 25985904

    Fluke extracellular vesicles were found not only in culture medium but in bile from infected hamsters and infected people, and when internalized drove cholangiocyte proliferation and interleukin-6 secretion; an anti-tetraspanin antibody blocked uptake. The human component is detection in bile, not an outcome measurement.

37. 37
    Thamavit W, Bhamarapravati N, Sahaphong S, Vajrasthira S, Angsubhakorn S. Effects of dimethylnitrosamine on induction of cholangiocarcinoma in Opisthorchis viverrini-infected Syrian golden hamsters.Cancer Res · 1978 · 38(12):4634-9PMID 214229

    The founding experiment of the field. No abstract exists in the PubMed record, so the incidence table and time points cannot be read from it; the result is verifiable only at second hand from the same group's later papers.

38. 38
    Thamavit W, Pairojkul C, Tiwawech D, Shirai T, Ito N. Strong promoting effect of Opisthorchis viverrini infection on dimethylnitrosamine-initiated hamster liver.Cancer Lett · 1994 · 78(1-3):121-5doi:10.1016/0304-3835(94)90040-x · PMID 8180954

    Quotes the 100% cholangiocarcinoma incidence from 1978 and attributes it to continuous nitrosamine dosing, then reports its own two-stage design: 44% cholangiocarcinoma and 93% cholangiofibrosis at 45 weeks, with the chemical alone producing hepatocellular nodules but no bile duct lesions. This is the source for the limit on the 100% figure.

39. 39
    Pairojkul C, Shirai T, Hirohashi S, et al. Multistage carcinogenesis of liver-fluke-associated cholangiocarcinoma in Thailand.Princess Takamatsu Symp · 1991 · 22:77-86PMID 1668894

    An independent restatement that all hamsters given subcarcinogenic nitrosamine plus flukes developed cholangiocarcinoma while neither agent alone did, with the human pathology beside it: among worm-positive people, only 25% showed adenomatous hyperplasia, the lesion thought to precede carcinoma. It also notes shared tumor phenotypes between fluke-related and non-fluke cancers, which later sequencing refined.

40. 40
    Srivatanakul P, Ohshima H, Khlat M, et al. Opisthorchis viverrini infestation and endogenous nitrosamines as risk factors for cholangiocarcinoma in Thailand.Int J Cancer · 1991 · 48(6):821-5doi:10.1002/ijc.2910480606 · PMID 1650329

    The human end of the nitrosamine argument: fluke-antibody-positive people in high-risk areas excreted 12.3 ± 18.7 µg of nitrosoproline per 12 hours after a proline load against 3.5 ± 3.2 in the negative, suppressed to 2.4 ± 2.0 by ascorbic acid, with small amounts of preformed nitrosamines found in fermented fish and pork. Area-level cancer incidence did not correlate with these measures.

41. 41
    Mitacek EJ, Brunnemann KD, Suttajit M, et al. Exposure to N-nitroso compounds in a population of high liver cancer regions in Thailand: volatile nitrosamine (VNA) levels in Thai food.Food Chem Toxicol · 1999 · 37(4):297-305doi:10.1016/s0278-6915(99)00017-4 · PMID 10418946

    Food chemistry rather than epidemiology: over 1,800 fresh and preserved food samples collected 1988-1996, with relatively high N-nitrosodimethylamine, N-nitrosopiperidine and N-nitrosopyrrolidine in fermented fish and up to 66.5 µg/kg of N-nitrosodimethylamine in salted and dried fish. It measures what is in the food, not what any individual ate.

42. 42
    Suyapoh W, Tangkawattana S, Suttiprapa S, Punyapornwithaya V, Tangkawattana P, Sripa B. Synergistic effects of cagA+ Helicobacter pylori co-infected with Opisthorchis viverrini on hepatobiliary pathology in hamsters.Acta Trop · 2020 · 213:105740doi:10.1016/j.actatropica.2020.105740 · PMID 33159904

    The fourth mechanism the standard three-strand account leaves out: the fluke carries Helicobacter pylori in its gut, and hamsters co-infected with the cagA-positive strain had higher inflammation scores, earlier and more severe dysplasia and a higher proliferation index, with co-infection and infection duration the strongest factors for periductal fibrosis. Hamster histopathology; no cancer incidence endpoint.

43. 43
    Miquel J, Świderski Z, Sripa B, Ribas A. Ultrastructural characters of the spermatozoon of the liver fluke Opisthorchis viverrini (Poirier, 1886) (Opisthorchiidae).Parasitol Res · 2017 · 116(9):2499-2506doi:10.1007/s00436-017-5559-y · PMID 28725936

    Cited for one thing only: the taxonomic authority carried in the species name, Poirier 1886, which fixes when the fluke was formally described. The paper itself is an electron-microscope study of the sperm cell and has nothing to do with cancer.

44. 44
    Schuster RK. Opisthorchiidosis--a review.Infect Disord Drug Targets · 2010 · 10(5):402-15doi:10.2174/187152610793180902 · PMID 20701569

    Source for the statement that the discovery history of the opisthorchiid flukes goes back more than 200 years, and for the older estimates of 17 million infected and more than 350 million at risk. A narrative review with no primary data.

45. 45
    Sadun EH. Studies on Opisthorchis viverrini in Thailand.Am J Hyg · 1955 · 62(2):81-115doi:10.1093/oxfordjournals.aje.a119772 · PMID 13258561

    The first substantial field study of this fluke in Thailand, cited here only for the date and the fact that the work was done. The PubMed record carries no abstract, so nothing in it can be quoted; its subject headings confirm only that it concerns Opisthorchis infection in Thailand.

46. 46
    Sadun EH, Chamnarnkit C, Chetanasen S. Studies on the treatment of Opisthorchis viverrini in human infections with quinacrine hydrochloride and chloroquine phosphate.Am J Trop Med Hyg · 1955 · 4(6):1080-7doi:10.4269/ajtmh.1955.4.1080 · PMID 13268815

    The first treatment trials against this infection, with quinacrine and chloroquine, a quarter of a century before praziquantel. The PubMed record has no abstract, so the cure rates cannot be read from it; the record establishes the date, the drugs and the fact that treatment was attempted.

47. 47
    Wykoff DE, Harinasuta C, Juttijudata P, Winn MM. Opisthorchis viverrini in Thailand--the life cycle and comparison with O. felineus.J Parasitol · 1965 · 51:207-14PMID 14275209

    The paper that established the life cycle in Thailand and compared it with the Russian species. No abstract exists in the PubMed record, so only its scope can be confirmed from the record: snails, fishes, seasons, classification and epidemiology.

48. 48
    Bunnag D, Harinasuta T. Studies on the chemotherapy of human opisthorchiasis in Thailand: I. Clinical trial of praziquantel.Southeast Asian J Trop Med Public Health · 1980 · 11(4):528-31PMID 7013095

    The arrival of the drug the control program now rests on: praziquantel cleared eggs from the stools of all 49 patients on two regimens, with no eggs found by day 60 and for four to eight months afterward. Mild transient side effects occurred in about 80% of the higher-dose group. Forty-nine patients, no control arm, and egg clearance as the only endpoint.

49. 49
    Bunnag D, Harinasuta T. Studies on the chemotherapy of human opisthorchiasis: III. Minimum effective dose of praziquantel.Southeast Asian J Trop Med Public Health · 1981 · 12(3):413-7PMID 7342331

    The dose-finding study behind the regimen used for control: 55 patients given a single 40 mg/kg dose had a 91% cure rate, against 88% for 25 mg/kg twice in one day and 44% for a single 25 mg/kg dose, and the authors recommend the single 40 mg/kg dose for treatment and control in endemic areas. Small groups, open label, cure measured by egg clearance at 30 and 60 days.

50. 50
    Jongsuksuntigul P, Imsomboon T. Opisthorchiasis control in Thailand.Acta Trop · 2003 · 88(3):229-32doi:10.1016/j.actatropica.2003.01.002 · PMID 14611877

    The administrative history of control: a small program from 1950, the three-part national strategy, and the reported positivity falling from 63.6% in 1984-87 to 35.6% in 1988 and 9.4% in 2001. Program figures, not a survey with a stated sampling frame.

51. 51
    Sithithaworn P, Haswell-Elkins MR, Mairiang P, et al. Parasite-associated morbidity: liver fluke infection and bile duct cancer in northeast Thailand.Int J Parasitol · 1994 · 24(6):833-43doi:10.1016/0020-7519(94)90009-4 · PMID 7982745

    The contemporary synthesis of the Thai evidence: about one third of the northeastern population infected, the parasite persisting despite praziquantel and education, and cholangiocarcinoma among the leading causes of death in the region. A review of geographic, hospital and community studies.

52. 52
    Flavell DJ, Lucas SB. Promotion of N-nitrosodimethylamine-initiated bile duct carcinogenesis in the hamster by the human liver fluke, Opisthorchis viverrini.Carcinogenesis · 1983 · 4(7):927-30doi:10.1093/carcin/4.7.927 · PMID 6307539

    The independent replication, with much lower yields: cholangiocarcinoma in 10% and 20% of the two combined groups, none in animals given the chemical alone or the parasite alone, and mean latencies of 249 and 308 days. It establishes that neither agent alone was sufficient in this design.

53. 53
    Parkin DM, Srivatanakul P, Khlat M, et al. Liver cancer in Thailand. I. A case-control study of cholangiocarcinoma.Int J Cancer · 1991 · 48(3):323-8doi:10.1002/ijc.2910480302 · PMID 1645697

    103 cases and matched controls: odds ratio 5.0 for raised antibody to the fluke, at least two-thirds of cases attributable, no association with hepatitis B, aflatoxin albumin adducts or dietary pattern, and a high risk in betel-nut users. Exposure is serological, so it cannot separate current from past infection.

54. 54
    Haswell-Elkins MR, Mairiang E, Mairiang P, et al. Cross-sectional study of Opisthorchis viverrini infection and cholangiocarcinoma in communities within a high-risk area in northeast Thailand.Int J Cancer · 1994 · 59(4):505-9doi:10.1002/ijc.2910590412 · PMID 7960220

    The dose-response, measured in a population: 12,311 adults from 85 villages stool-graded, 1,807 scanned, 15 preclinical cancers, and a sharply elevated adjusted prevalence odds ratio of 14.1 in the most heavily infected group. The cancer endpoint is ultrasound-based with endoscopic confirmation where possible, not uniformly histological.

55. 55
    Ittiprasert W, Mann VH, Karinshak SE, et al. Programmed genome editing of the omega-1 ribonuclease of the blood fluke, Schistosoma mansoni.eLife · 2019 · 8:e41337doi:10.7554/eLife.41337 · PMID 30644357

    The companion paper to the granulin knockout, published the same day, 15 January 2019, in the same journal and with overlapping authors. Its first line is that CRISPR/Cas9 genome editing “has yet to be reported in species of the Platyhelminthes,” which is why neither paper can be called the first on its own. It edited a blood fluke, not a liver fluke, and its endpoints were immune polarization and granuloma volume, not cancer. The PubMed title is truncated where the species name was italicized.

56. 56
    Steele JA, Richter CH, Echaubard P, et al. Thinking beyond Opisthorchis viverrini for risk of cholangiocarcinoma in the lower Mekong region: a systematic review and meta-analysis.Infect Dis Poverty · 2018 · 7(1):44doi:10.1186/s40249-018-0434-3 · PMID 29769113

    The most uncomfortable finding in this literature: pooling 18 studies, alcohol combined with smoking gave an odds ratio of 11.1 (95% CI 5.63-21.92), which the authors describe as a greater risk factor than exposure to the fluke; family history, raw cyprinoid fish, high-nitrate foods and praziquantel treatment were each also associated with increased risk. Heterogeneous observational studies, random-effects pooling.

57. 57
    Shin HR, Oh JK, Masuyer E, et al. Epidemiology of cholangiocarcinoma: an update focusing on risk factors.Cancer Sci · 2010 · 101(3):579-85doi:10.1111/j.1349-7006.2009.01458.x · PMID 20085587

    Puts the liver flukes beside the viral causes: summary relative risks of 4.8 (95% CI 2.8-8.4) for liver fluke infection, 2.6 for hepatitis B and 1.8 for hepatitis C, and an average O. viverrini prevalence in Thailand of 9.6%, about 6 million people, at that time. A meta-analysis of observational studies, written by an IARC author.

58. 58
    Ong CK, Subimerb C, Pairojkul C, et al. Exome sequencing of liver fluke-associated cholangiocarcinoma.Nat Genet · 2012 · 44(6):690-3doi:10.1038/ng.2273 · PMID 22561520

    The first exome sequencing of fluke-associated bile duct cancer: eight tumors with matched normal tissue sequenced in full, 206 validated somatic mutations in 187 genes, and 15 genes screened in 46 further patients. TP53 was mutated in 44.4% of cases, KRAS and SMAD4 in 16.7% each, and ten genes not previously implicated were hit in 3.7% to 14.8%, among them MLL3, ROBO2, RNF43, PEG3 and GNAS. Eight tumors is a very small discovery set, and there is no comparison group of non-fluke cancers in this paper.

59. 59
    Chan-On W, Nairismägi ML, Ong CK, et al. Exome sequencing identifies distinct mutational patterns in liver fluke-related and non-infection-related bile duct cancers.Nat Genet · 2013 · 45(12):1474-8doi:10.1038/ng.2806 · PMID 24185513

    209 cholangiocarcinomas, 108 fluke-related: BAP1, IDH1 and IDH2 were mutated more often in the non-fluke cancers and TP53 showed the reciprocal pattern, establishing that different causes leave different mutations in the same tumor type. Sequencing of tumor tissue; no clinical outcomes.

60. 60
    Kamsa-ard S, Laopaiboon M, Luvira V, Bhudhisawasdi V. Association between praziquantel and cholangiocarcinoma in patients infected with Opisthorchis viverrini: a systematic review and meta-analysis.Asian Pac J Cancer Prev · 2013 · 14(11):7011-6doi:10.7314/apjcp.2013.14.11.7011 · PMID 24377641

    The pooled estimate that came before the 2015 case-control study and did not find the association: three studies and 637 patients were included, and the pooled odds ratio, computed from the two studies of 237 patients that could be combined, was 1.8 (95% CI 0.81–4.16). The authors call their own evidence inconclusive and name the methodological limits, in particular the lack of histological confirmation.

61. 61
    Hanpanich P, Laha T, Sripa B, et al. Decreased risk of cholangiocarcinogenesis following repeated cycles of Opisthorchis viverrini infection-praziquantel treatment: Magnetic Resonance Imaging (MRI) and histopathological study in a hamster model.Parasitol Int · 2017 · 66(4):464-470doi:10.1016/j.parint.2016.04.012 · PMID 27143606

    The animal test of the claim that repeated infection and repeated praziquantel raise cancer risk, and it found the opposite: hamsters given three rounds of infection each followed by praziquantel developed no cholangiocarcinoma by week 40, while the groups given fluke plus nitrosamine, or nitrosamine alone, did. Twenty-five hamsters in five groups, so five animals per group — too few to measure a difference, enough to undercut the premise the human studies were built on.

62. 62
    Worasith C, Techasen A, Duenngai K, et al. Comparison of a Urine Antigen Assay and Multiple Examinations with the Formalin-Ethyl Acetate Concentration Technique for Diagnosis of Opisthorchiasis.Am J Trop Med Hyg · 2023 · 109(1):159-165doi:10.4269/ajtmh.23-0132 · PMID 37217166

    The best independent check on the urine antigen result, because it looks harder with the old method instead of trusting the new one: reading five drops of fecal sediment instead of the usual two raised sensitivity from 58.2% to 67% and found eggs in 25 of 82 people (30.5%) who were antigen-positive and egg-negative on the standard protocol, and in 2 of 80 antigen-negative people. It cannot rule out that some antigen positives are still false, and the reference standard again combines both tests.

63. 63
    Worasith C, Sithithaworn J, Wongphutorn P, et al. Accuracy of a new rapid diagnostic test for urinary antigen detection and assessment of drug treatment in opisthorchiasis.Infect Dis Poverty · 2023 · 12(1):102doi:10.1186/s40249-023-01162-4 · PMID 37990282

    The strip version of the urine antigen test, and the only place on this page where a specificity is actually reported: 94.2% sensitivity and 93.2% specificity against stool concentration in 493 people, 2% cross-reaction with other worms, 53.2% against 54.0% prevalence beside the laboratory assay in a field trial of 1,629, and a one-month cure rate of 86.6% after praziquantel. The comparison standard is stool examination, which the same group shows misses light infections, so the specificity is measured against an imperfect reference.

64. 64
    Heneberg P. From infection to cholangiocarcinoma: Why opisthorchiids break these rules.One Health · 2026 · 23:101517doi:10.1016/j.onehlt.2026.101517 · PMID 42500360

    The current conceptual reframing: it argues that chronic inflammation is not intrinsically carcinogenic, proposes a pro-oncogenic inflammation threshold requiring several dimensions to converge, and explains why most helminths stay below it while the opisthorchiids exceed it. A review that synthesizes rather than measures.

65. 65
    Khounvisith V, Virachith S, Innoula N, et al. High infection risk of intestinal helminths despite WASH interventions: A cross-sectional study in Khammouane province, Lao PDR.PLoS Negl Trop Dis · 2026 · 20(6):e0014388doi:10.1371/journal.pntd.0014388 · PMID 42224275

    The current picture in Laos, measured by stool concentration in 1,530 randomly chosen adults across 81 villages where water and sanitation programs had been run: O. viverrini in 56.9%, hookworm in 54.7%, minute intestinal flukes in 22.7%, and no association between better water or sanitation and less infection. Cross-sectional, and measured by the egg method this page shows underestimates light infection, so the figure is a floor.

66. 66
    Rodpai R, Boonroumkaew P, Sadaow L, et al. Co-endemicity of Opisthorchis viverrini and Strongyloides stercoralis infections: evidence from stool examination and serology screening in Champasak Province, Southern Lao People's Democratic Republic.Parasitol Res · 2026 · 125(1)doi:10.1007/s00436-026-08698-y · PMID 42185667

    The same gap between tests, in southern Laos: among 242 adults in three villages, 51.7% were positive by stool concentration and 65.3% by an antibody strip, with an estimated true prevalence of 66.6% once both were combined. Two hundred and forty-two people in three villages, and the antibody test cannot distinguish current from past infection, so the combined estimate is an upper bound.

67. 67
    Wattanawong O, Iamsirithaworn S, Kophachon T, et al. Current status of helminthiases in Thailand: A cross-sectional, nationwide survey, 2019.Acta Trop · 2021 · 223:106082doi:10.1016/j.actatropica.2021.106082 · PMID 34364893

    Thailand's national survey: 16,187 stool samples from 76 provinces, overall helminth prevalence 9.79%, O. viverrini 2.2% with a mean of 120.9 eggs per gram and 99.1% of infections low intensity. Everything here is measured by microscopy, which is the limitation the later antigen work exposes.

68. 68
    Sodamuk P, Patthaisong W, Hassan NW, Sangaran K, Chuangchaiya S, Idris ZM. Molecular evidence of a high prevalence of Opisthorchis viverrini infection in an endemic community in Sakon Nakhon Province, northeastern Thailand.Trop Med Health · 2026 · 54(1)doi:10.1186/s41182-026-00929-3 · PMID 41723506

    A 2025 community survey in which microscopy found 6.1% infected, every case very light at under 500 eggs per gram, while ITS2 PCR found 19.9% - more than three times as many - with male sex carrying an adjusted odds ratio of 5.55. 246 participants in one district.

69. 69
    Rodpai R, Janwan P, Sadaow L, et al. Current Prevalence of Infection with Opisthorchis viverrini and Intestinal Parasites in Rural Communities in Northeast Thailand Using Stool Examinations and Serological Screening Methods.Am J Trop Med Hyg · 2025 · 113(4):821-826doi:10.4269/ajtmh.25-0258 · PMID 40763723

    In 307 rural participants, stool concentration found liver fluke eggs in 16% while an immunochromatographic strip was positive in 26.1% and ELISA in 31.6%, a statistically significant difference. Antibody positivity does not distinguish current from past infection, and the abstract does not raise the point; the authors propose these as screening tools.

70. 70
    Florio AA, Ferlay J, Znaor A, et al. Global trends in intrahepatic and extrahepatic cholangiocarcinoma incidence from 1993 to 2012.Cancer · 2020 · 126(11):2666-2678doi:10.1002/cncr.32803 · PMID 32129902

    Registry-based global comparison: the highest intrahepatic rates were in South Korea (2.80), Thailand (2.19) and Japan (0.95), with incidence of both subtypes rising in most of 38 countries and the largest extrahepatic rise in Thailand at 8.8% a year. These are national rates, so they average the Mekong basin with the rest of each country.

71. 71
    Homsana A, Southisavath P, Kling K, et al. Burden and risk factors of suspected cholangiocarcinoma in high Opisthorchis viverrini endemic rural communities in southern Lao PDR.PLoS Negl Trop Dis · 2024 · 18(11):e0012617doi:10.1371/journal.pntd.0012617 · PMID 39602377

    Current large field evidence from Laos: among 3,400 people, 88.3% were infected and 7.2% (95% CI 5.4-9.6) had suspected cholangiocarcinoma, with infection carrying an adjusted odds ratio of 3.4. The endpoint is ultrasound-suspected cancer, not histologically confirmed cancer, which makes the prevalence figure a surrogate.

72. 72
    Moungthard H, Thinkhamrop K, Chamadol N, Laopaiboon V, Thinkhamrop B, Kelly M. Association of Cholangiocarcinoma among People with Periductal Fibrosis.Asian Pac J Cancer Prev · 2023 · 24(12):4147-4154doi:10.31557/APJCP.2023.24.12.4147 · PMID 38156850

    The grading of the screening marker against outcome in 751,061 participants: periductal fibrosis prevalence 15.35%, cancer 0.11%, with adjusted odds ratios of 0.94 for peripheral duct fibrosis, 1.4 for segmental and 2.52 for main duct. Cross-sectional within a screening program, with fibrosis read by trained general practitioners or radiologists.

73. 73
    Lovis L, Mak TK, Phongluxa K, et al. PCR Diagnosis of Opisthorchis viverrini and Haplorchis taichui Infections in a Lao Community in an area of endemicity and comparison of diagnostic methods for parasitological field surveys.J Clin Microbiol · 2009 · 47(5):1517-23doi:10.1128/JCM.02011-08 · PMID 19279176

    The paper that quantifies both the minute-intestinal-fluke confusion and the sampling problem: species-specific PCR found 93.7% of liver fluke infections against purge results, community prevalence was 63.9% for the liver fluke and 30.6% for Haplorchis taichui, and against PCR three Kato-Katz smears found 91.4% of infections while one smear found 62.3% and the concentration technique 49.3%. One Lao community.

74. 74
    Prakobwong S, Charoensuk L, Chedtabud K, Pinlaor S, Poonlaphdecha S, Ribas A. Enhanced detection of Opisthorchis viverrini infection: A comparative evaluation of modified one-step FECT and conventional diagnostic methods in low-intensity setting.Parasite Epidemiol Control · 2024 · 27:e00391doi:10.1016/j.parepi.2024.e00391 · PMID 39687917

    A modified concentration technique reached 95.6% sensitivity against 87.9% for the conventional version, 55.5% for Kato-Katz and 42.3% for a simple smear, with a detection limit of 18 against 34 eggs per gram; in 3,900 field samples, 67% of those below 50 eggs per gram were missed by the conventional technique. A method study, not a prevalence survey.

75. 75
    Thinkhamrop K, Khuntikeo N, Laohasiriwong W, Chupanit P, Kelly M, Suwannatrai AT. Association of comorbidity between Opisthorchis viverrini infection and diabetes mellitus in the development of cholangiocarcinoma among a high-risk population, northeastern Thailand.PLoS Negl Trop Dis · 2021 · 15(9):e0009741doi:10.1371/journal.pntd.0009741 · PMID 34543283

    In 263,776 screened people, cancer was found in 0.47% of those reporting fluke infection, 0.59% of those with diabetes and 0.73% of those with both, with an adjusted odds ratio of 2.36 (95% CI 1.74-3.21) for the combination against neither. Infection and diabetes were self-reported on a questionnaire.

76. 76
    Wangboon C, Yongvanit P, Loilome W, et al. Elevated Levels of Urinary 8-oxodG Correlate with Persistent Periductal Fibrosis after Praziquantel Treatment in Chronic Opisthorchiasis.Am J Trop Med Hyg · 2018 · 98(6):1763-1769doi:10.4269/ajtmh.17-0971 · PMID 29637887

    The human answer to whether killing the fluke undoes the damage: urinary oxidative DNA damage fell after treatment in the 14 infected people without fibrosis and was unchanged in the 52 with it; 65.3% of those with fibrosis became ultrasound-negative a year later while 34.6% did not. Small numbers, one year of follow-up, no cancer endpoint.

77. 77
    Vale N, Gouveia MJ, Botelho M, et al. Carcinogenic liver fluke Opisthorchis viverrini oxysterols detected by LC-MS/MS survey of soluble fraction parasite extract.Parasitol Int · 2013 · 62(6):535-42doi:10.1016/j.parint.2013.08.001 · PMID 23973383

    Mass spectrometry of adult fluke extracts indicating novel oxysterol derivatives, several compatible with an estrogen core and with oxidation to an estradiol-2,3-quinone. It identifies chemistry in the parasite; it does not show that chemistry damages human DNA in a person.

78. 78
    Jusakul A, Loilome W, Namwat N, et al. Liver fluke-induced hepatic oxysterols stimulate DNA damage and apoptosis in cultured human cholangiocytes.Mutat Res · 2012 · 731(1-2):48-57doi:10.1016/j.mrfmmm.2011.10.009 · PMID 22044627

    Five oxysterols were identified in livers of infected hamsters; two were significantly higher in animals with cholangiocarcinoma, and both induced apoptosis and three DNA adducts, including 8-oxo-7,8-dihydro-2'-deoxyguanosine, in cultured human cholangiocytes. Animal plus cell culture; the link to human cancer is proposed.

79. 79
    Brindley PJ, Correia da Costa JM, Sripa B. Why does infection with some helminths cause cancer?Trends Cancer · 2015 · 1(3):174-182doi:10.1016/j.trecan.2015.08.011 · PMID 26618199

    The conceptual paper behind the metabolite hypothesis: it asks why only a few trematodes are Group 1 carcinogens and proposes catechol-estrogen and oxysterol metabolites of parasite origin as initiators. An interpretation, offered as such by its authors.

80. 80
    Arunsan P, Chaidee A, Cochran CJ, et al. Liver fluke granulin promotes extracellular vesicle-mediated crosstalk and cellular microenvironment conducive to cholangiocarcinoma.Neoplasia · 2020 · 22(5):203-216doi:10.1016/j.neo.2020.02.004 · PMID 32244128

    This is the paper that knocked out the HOST gene, human progranulin, and only in a cultured human cholangiocyte line, to read the fluke protein's effect without the host's own version present. It is almost certainly the source of the common claim that a human gene was edited in the hamster experiments. In vitro only.

81. 81
    Chaiyadet S, Ittiprasert W, Smout MJ, et al. Gene Editing of a Carcinogenic Liver Fluke Tetraspanin Impairs Parasite Surface Biogenesis and Extracellular Vesicle Uptake by Human Host Cells.J Infect Dis · 2026 · 233(2):e510-e520doi:10.1093/infdis/jiaf466 · PMID 40905941

    The most current primary paper in this line: tetraspanin knockout produced a vacuolated surface, far fewer secreted vesicles and poor uptake by human cholangiocytes. It also states that Ov-GRN-1 is not detected in extracellular vesicles and that granulin knockout is not lethal in vivo, which disciplines the granulin story. The authors advocate this antigen as a vaccine, which is their own program. The two statements about Ov-GRN-1 — that it is not detected in extracellular vesicles and that knocking out its gene is not lethal to the fluke in vivo — are in the Methods of the full text (PMC13016728), not the abstract. The paper says nothing about heritable transgenic lines.

82. 82
    Jusakul A, Cutcutache I, Yong CH, et al. Whole-Genome and Epigenomic Landscapes of Etiologically Distinct Subtypes of Cholangiocarcinoma.Cancer Discov · 2017 · 7(10):1116-1135doi:10.1158/2159-8290.CD-17-0368 · PMID 28667006

    489 cholangiocarcinomas from 10 countries, 71 of them whole-genome sequenced: integrative clustering produced four clusters that split by cause rather than anatomy, with distinct methylation patterns, and the authors conclude that molecular landscapes differ radically by etiology. It does not test whether that difference should change treatment.

83. 83
    Sitthirak S, Wangwiwatsin A, Jusakul A, et al. Whole exome sequencing of multi-regions reveals tumor heterogeneity in Opisthorchis viverrini-associated cholangiocarcinoma.Sci Rep · 2025 · 15(1):10886doi:10.1038/s41598-025-95142-3 · PMID 40157958

    52 tumor samples and 13 adjacent tissues from 13 patients: 48% to 90% of mutations were confined to a single region of the tumor, with an average of 95 non-synonymous mutations per region and recurrent TP53 and SMAD4 drivers. Thirteen patients, so the heterogeneity is better established than its frequency.

84. 84
    Deenonpoe R, Guscott MA, Watcharadetwittaya S, et al. Chromosomal instability and genomic alterations in cholangiocarcinoma from Northeastern Thailand.J Pathol · 2025 · 267(3):261-274doi:10.1002/path.6464 · PMID 40952339

    Fluke-positive cholangiocarcinoma cell lines showed elevated chromosome segregation errors against a normal cholangiocyte line, though at lower rates than colorectal or ovarian cancer, and 33 tumors with matched normal tissue showed whole and subchromosomal alterations including CCNE1 amplification and one rare BRCA1 amplification. Cell lines plus low-pass sequencing; no therapeutic test.

85. 85
    Plum PS, Hess T, Bertrand D, et al. Integrative genomic analyses of European intrahepatic cholangiocarcinoma: Novel ROS1 fusion gene and PBX1 as prognostic marker.Clin Transl Med · 2024 · 14(6):e1723doi:10.1002/ctm2.1723 · PMID 38877653

    The European contrast case: 37 German intrahepatic cholangiocarcinomas with extended screening of 40 more, finding PBX1 copy gain in 20% of tumors, which the authors note tends to occur in European rather than liver-fluke-associated Asian disease, along with novel fusion genes. A different population, not a comparison study of fluke status.

86. 86
    Oura K, Morishita A, Nakahara M, et al. Chronic Liver Disease Associated Cholangiocarcinoma: Genomic Insights and Precision Therapeutic Strategies.Cancers (Basel) · 2025 · 17(18):3052doi:10.3390/cancers17183052 · PMID 41008893

    A current review of how cause shapes the molecular landscape, stating that liver fluke infection and primary sclerosing cholangitis are associated with distinct signatures including TP53, SMAD4, KRAS and ERBB2 alterations, and that targetable alterations such as FGFR2 fusions, IDH1 mutations and ERBB2 amplification are heterogeneously distributed across etiologies and anatomical subtypes. It calls for etiology-stratified genomic studies, which is to say they do not yet exist.

87. 87
    Erul E, Cifuentes-Canaval S, Santhosh A, et al. Fibroblast Growth Factor Receptor (FGFR) Inhibitors for the Treatment of Cholangiocarcinoma: Key Therapeutic Developments and Knowledge Gaps.Drug Des Devel Ther · 2026 · 20:559328doi:10.2147/DDDT.S559328 · PMID 41773262

    The current review of the drug class aimed at FGFR2, cited here for one statement: that the prevalence of FGFR2 fusions and mutations varies by region and by cause, liver fluke-associated disease among them. A narrative review of trial results; it gives no fluke-stratified response figures, which is the gap this page is pointing at.

88. 88
    Woo S, Kim Y, Hwang S, Chon HJ. Epidemiology and genomic features of biliary tract cancer and its unique features in Korea.J Liver Cancer · 2025 · 25(1):41-51doi:10.17998/jlc.2025.02.27 · PMID 40033637

    The source for which targeted drugs are approved and for whom: ivosidenib for IDH1 mutations and pemigatinib for FGFR2 fusions, both approved in Korea, in a population where liver fluke infection and hepatitis B are the distinctive causes and where intrahepatic tumors carry lower frequencies of IDH1 mutation and FGFR2 fusion than elsewhere. A regional review, so the approvals it describes are Korean and the frequency comparisons are not a fluke-stratified analysis.

89. 89
    Charoensuk L, Subrungruang I, Mungthin M, Pinlaor S, Suwannahitatorn P. Comparison of stool examination techniques to detect Opisthorchis viverrini in low intensity infection.Acta Trop · 2019 · 191:13-16doi:10.1016/j.actatropica.2018.12.018 · PMID 30550733

    Head-to-head comparison of routine methods: the formalin-ethyl acetate concentration technique gave the highest sensitivity at 91.0%, including 94.3% below 50 eggs per gram, ahead of Kato-Katz, a commercial concentrator kit and a direct smear, and the authors explicitly do not recommend substituting the kit for light infections.

90. 90
    Jongthawin J, Mahittikorn A, Wangdi K, Masangkay FR, Kotepui M. Comparison of Ritchie and Kato-Katz methods for the detection of intestinal helminths in humans: a systematic review and meta-analysis.Parasit Vectors · 2026 · 19(1)doi:10.1186/s13071-026-07437-7 · PMID 42141455

    The pooled answer to which routine stool method is better, across 40 studies: for O. viverrini there was no difference between formalin-ether concentration and a single Kato-Katz smear, odds ratio 1.09 (95% CI 0.61–1.92), with very high heterogeneity between studies (I² 86.7%). Heterogeneity that high means the pooled figure hides real differences between settings, and the review does not report how many of the 40 studies contributed the liver fluke comparison.

91. 91
    Phuphisut O, Poodeepiyasawat A, Yoonuan T, et al. Ov-RPA-CRISPR/Cas12a assay for the detection of Opisthorchis viverrini infection in field-collected human feces.Parasit Vectors · 2024 · 17(1):80doi:10.1186/s13071-024-06134-7 · PMID 38383404

    The clearest measurement of how often microscopy names the wrong fluke: of 29 field samples reported as carrying O. viverrini-like eggs, a DNA assay using CRISPR as the detector confirmed the liver fluke in 18, leaving 11 that the authors read as most likely minute intestinal flukes. The assay is a method paper with a 121-sample validation set, and it is a confirmatory test, not a prevalence survey.

92. 92
    Junsawang P, Techasen A, Wiratchawa K, et al. AI-enhanced rapid diagnostic testing platform for mass opisthorchiasis screening.Sci Rep · 2025 · 15(1):31062doi:10.1038/s41598-025-16893-7 · PMID 40849358

    How the urine strip is being read at scale: a phone application and cloud service that grades the test line, reported as 98% accuracy for image quality and 95% for infection status, and used on more than 100,000 samples in northeastern Thailand. The authors state plainly that performance elsewhere has not been validated. This is a measurement of the reader against human grading, not of the test against infection.

93. 93
    Homsana A, Odermatt P, Southisavath P, Yajima A, Sayasone S. Cross-reaction of POC-CCA urine test for detection of Schistosoma mekongi in Lao PDR: a cross-sectional study.Infect Dis Poverty · 2020 · 9(1):114doi:10.1186/s40249-020-00733-z · PMID 32787912

    Cited here for one practical finding: positive results on the point-of-care circulating cathodic antigen strip, designed for schistosomiasis, were significantly associated with Opisthorchis viverrini infection (1.69, 95% CI 1.02-2.77), so the strip can report the wrong parasite where both occur.

94. 94
    Khuntikeo N, Chamadol N, Yongvanit P, et al. Cohort profile: cholangiocarcinoma screening and care program (CASCAP).BMC Cancer · 2015 · 15:459doi:10.1186/s12885-015-1475-7 · PMID 26054405

    The design paper for the screening program this page relies on: a prospective screening cohort intended to exceed 150,000 people with annual ultrasound, plus a patient cohort, with 85,927 enrolled and 55.0% scanned at the time of the report. A cohort profile, so it reports enrollment rather than outcomes.

95. 95
    Khuntikeo N, Titapun A, Loilome W, et al. Current Perspectives on Opisthorchiasis Control and Cholangiocarcinoma Detection in Southeast Asia.Front Med (Lausanne) · 2018 · 5:117doi:10.3389/fmed.2018.00117 · PMID 29765958

    The program's own statement of its design, and the source for the fact that it now screens for the worm with the urine antigen assay as primary prevention and for periductal fibrosis and cancer with ultrasound as secondary prevention. A review by the program's architects; it sets out intentions and anticipated impacts rather than measuring them.

96. 96
    Prathumkam P, Thinkhamrop K, Khuntikeo N, Chamadol N, Thuanman J, Kelly M, Thinkhamrop B. Association between the Number of Repeated Praziquantel Treatments and Kidney Parenchymal Change in Northeast Thailand.Asian Pac J Cancer Prev · 2022 · 23(7):2397-2405doi:10.31557/APJCP.2022.23.7.2397 · PMID 35901347

    In 490,969 screened people, kidney abnormality on ultrasound rose with the number of praziquantel courses, reaching an odds ratio of 1.25 (95% CI 1.02-1.52) above three courses. Treatment history was self-reported, the absolute prevalence difference is small, and the authors judge the drug's benefit to override it for infected people.

97. 97
    Kovner A, Kapushchak Y, Pakharukova M. Liver flukes and kidney injury: systematic review of human and animal data (from 1950 to 2025).J Helminthol · 2026 · 100:e24doi:10.1017/S0022149X26101187 · PMID 41766340

    The reason a kidney finding in treated people cannot be read as a drug effect without care: this review of human and animal data from 1950 to 2025 concludes that liver fluke infection itself is associated with kidney damage — glomerular and tubular injury, interstitial inflammation and fibrosis — through immune complexes and oxidative stress. It pools studies of several fluke species and very mixed design, and it reports no effect size.

98. 98
    Thamavit W, Moore MA, Sirisinha S, Shirai T, Ito N. Time-dependent modulation of liver lesion development in Opisthorchis-infected Syrian hamster by an antihelminthic drug, praziquantel.Jpn J Cancer Res · 1993 · 84(2):135-8doi:10.1111/j.1349-7006.1993.tb02846.x · PMID 8463130

    Praziquantel given 4, 12 or 20 weeks after infection reduced bile duct lesions significantly only at 4 weeks, which the authors read as showing that the fluke's enhancement of bile duct carcinogenesis is rapid and largely irreversible. An animal result with a chemical initiator, not a human treatment study.

99. 99
    Kopolrat KY, Boueroy P, Kammoolkon R, et al. Longitudinal changes and risk factors of Opisthorchis viverrini infection after selective praziquantel treatment: evidence from urine antigen assay and fecal examination in an endemic community in Northeast Thailand.PLoS One · 2026 · 21(7):e0352854doi:10.1371/journal.pone.0352854 · PMID 42406772

    The clearest reinfection measurement, and it depends entirely on the test: in 612 people over 24 weeks, reinfection after praziquantel ran at 63.7 per 100 person-years by urine antigen against 5.9 by stool examination, with raw fish consumption in the previous six months carrying an adjusted risk ratio of 7.52. A single community; the two tests disagree about who is infected.

100. 100
     Saengsawang P, Buakate P. Re-Infection Rate of Opisthorchis Viverrini Five Years After Treatment with Praziquantel in High-Risk Area: A Community-Based Study.Asian Pac J Cancer Prev · 2024 · 25(8):2679-2684doi:10.31557/APJCP.2024.25.8.2679 · PMID 39205565

     The long-interval counterpart: of 418 people followed five years after treatment and examined by stool concentration, 10.8% were reinfected. Measured by the less sensitive method, so this is a floor rather than a rate.

101. 101
     Sripa B, Tangkawattana S, Sangnikul M. Integrated opisthorchiasis control through the EcoHealth/one health approach: 15 years of success and experiences with the Lawa model.One Health · 2025 · 20:101006doi:10.1016/j.onehlt.2025.101006 · PMID 40123921

     The program's own fifteen-year account: infection around Lawa Lake down from 60% to under 5%, infection in the snail and cyprinid fish hosts now undetectable, and the model turned into a training hub being scaled up in Thailand and the Mekong region. Written by the people who built it, with no independent evaluation and no confidence intervals or sample sizes in the abstract.

102. 102
     Sripa B, Tangkawattana S, Sangnikul T. The Lawa model: A sustainable, integrated opisthorchiasis control program using the EcoHealth approach in the Lawa Lake region of Thailand.Parasitol Int · 2017 · 66(4):346-354doi:10.1016/j.parint.2016.11.013 · PMID 27890720

     The fullest account of the control program: infection in the 12 villages around the lake below 10% from an average of 60%, cyprinid fish under 1% from a maximum of 70%, and no infected Bithynia found in recent surveys, against a national backdrop of up to 85% prevalence persisting in some endemic areas. It is an evaluation of the authors' own intervention, with no independent assessment.

103. 103
     Sripa B, Tangkawattana S, Laha T, Kaewkes S, Mallory FF, Smith JF, Wilcox BA. Toward integrated opisthorchiasis control in northeast Thailand: the Lawa project.Acta Trop · 2015 · 141(Pt B):361-7doi:10.1016/j.actatropica.2014.07.017 · PMID 25102053

     The earlier report of the same program, recording infection in the surrounding villages falling to approximately one third of the 50% baseline average and fish infection under 1% against a maximum of 70%. Same conflict of interest as the later paper.

104. 104
     Chamadol N, Laopaiboon V, Jareanrat A, et al. Improvement of survival outcomes of cholangiocarcinoma by ultrasonography surveillance: Multicenter retrospective cohorts.Heliyon · 2024 · 10(18):e38191doi:10.1016/j.heliyon.2024.e38191 · PMID 39381227

     In 711 histologically proven cases, five-year survival was 53.9% with a median of 67.2 months in the screened group against 21.9% and 15.6 months in people presenting with symptoms. Retrospective and not randomized, so part of the difference is earlier diagnosis of the same disease rather than longer life.

105. 105
     Thanasukarn V, Srisuk T, Luvira V, et al. Improving postoperative survival in cholangiocarcinoma: development of surgical strategies with a screening program in the epidemic region.World J Surg Oncol · 2024 · 22(1):287doi:10.1186/s12957-024-03573-5 · PMID 39478620

     A single-center series of 1,091 resected patients comparing 2002-2013 with 2014-2021: early-stage disease rose from 16.0% to 29.1%, positive surgical margins fell from 53.7% to 40.0%, and median survival rose from 14 to 40 months. Surgical technique changed over the same period, so the screening program is not the only variable.

106. 106
     O'Connor SY, Mationg ML, Kelly MJ, et al. The Impact of "The Magic Glasses Opisthorchiasis" on Schoolchildren's Knowledge, Attitudes and Practices Surrounding Opisthorchis viverrini in the Lower Mekong Basin, a Cluster-Randomised Controlled Trial.Trop Med Infect Dis · 2026 · 11(7):174doi:10.3390/tropicalmed11070174 · PMID 42506744

     A cluster-randomized trial of a cartoon-based school education package in Cambodia, Laos and Thailand: knowledge and attitude scores rose by 19.2 and 25.3 percentage points in Cambodia and 19.0 and 14.2 in Laos, but fell by 23.3 and 15.8 in Thailand, and behavior scores improved in no country. The outcome is a questionnaire score, not infection.

107. 107
     Mationg ML, Clements ACA, Williams GM, et al. A study protocol for a multi-country cluster randomized controlled trial of the impact of a multi-component One Health strategy to eliminate Opisthorchis viverrini and soil transmitted helminths in the Lower Mekong Basin.Trials · 2024 · 25(1):801doi:10.1186/s13063-024-08616-6 · PMID 39605028

     The first randomized test of the Lawa approach, and so far only a plan: 18 rural communities across Cambodia, Laos and Thailand randomized to one round of praziquantel and albendazole, or to that plus the Lawa model and the Magic Glasses education package, with cumulative incidence of infection at 12 months as the primary outcome, measured by concentration technique and quantitative PCR. Registered as ACTRN12622000353796. A protocol carries no results, and its authors include the program's founder.

108. 108
     Ziegler AD, Echaubard P, Lee YT, et al. Untangling the Complexity of Liver Fluke Infection and Cholangiocarcinoma in NE Thailand Through Transdisciplinary Learning.EcoHealth · 2016 · 13(2):316-27doi:10.1007/s10393-015-1087-3 · PMID 26822781

     A field study with 72 students in five villages whose central argument is methodological: viewing liver fluke infection and cholangiocarcinoma as the same issue is inadvisable, because multiple factors influence each, and an incomplete picture of the linkages may reduce the efficacy of interventions. Qualitative work; it generates no effect estimates.

109. 109
     Chaiyadet S, Sotillo J, Krueajampa W, et al. Vaccination of hamsters with Opisthorchis viverrini extracellular vesicles and vesicle-derived recombinant tetraspanins induces antibodies that block vesicle uptake by cholangiocytes and reduce parasite burden after challenge infection.PLoS Negl Trop Dis · 2019 · 13(5):e0007450doi:10.1371/journal.pntd.0007450 · PMID 31136572

     Vaccinated hamsters recovered significantly fewer and shorter adult flukes, with egg output falling significantly in only one group, and antibodies blocked vesicle uptake by human cholangiocytes in vitro. The authors describe the efficacy as partial; there is no human vaccine work.

110. 110
     Smout MJ, Sripa B, Laha T, et al. Infection with the carcinogenic human liver fluke, Opisthorchis viverrini.Mol Biosyst · 2011 · 7(5):1367-75doi:10.1039/c0mb00295j · PMID 21311794

     The source of the much-repeated framing that more ubiquitous carcinogenic microbes such as Helicobacter pylori induce cancer in less than 1% of infected people while as many as one sixth of people with opisthorchiasis will develop cholangiocarcinoma. It is a review assertion, not a cohort measurement, and it does not reconcile with registry incidence rates.

111. 111
     Anchalee N, Thinkhamrop K, Suwannatrai AT, Titapun A, Loilome W, Kelly M. Spatio-Temporal Analysis of Cholangiocarcinoma in a High Prevalence Area of Northeastern Thailand: A 10-Year Large Scale Screening Program.Asian Pac J Cancer Prev · 2024 · 25(2):537-546doi:10.31557/APJCP.2024.25.2.537 · PMID 38415540

     The one large recent dataset that points the other way from the Khon Kaen registry: 2,515 cases confirmed by pathology in the screening program across 20 provinces and 282 districts between 2013 and 2022, described as continuing to rise and to spread outward from hotspots in the upper and middle northeast. These are counts inside a program that finds more cancer the more it screens, not an age-standardized population rate, so it cannot be compared directly with registry trends.

112. 112
     Luvira V, Kamsa-Ard S, Kamsa-Ard S, et al. Association between Repeated Praziquantel treatment and Papillary, and Intrahepatic Cholangiocarcinoma.Ann Hepatol · 2018 · 17(5):802-809doi:10.5604/01.3001.0012.3140 · PMID 30145559

     The same design split by tumor type: the association with repeated praziquantel held for intrahepatic cancers (odds ratio 4.21, 95% CI 1.61-11.05 for three or more courses) and papillary cancers, and not for extrahepatic or tubular ones. Same confounding caveat as the parent study.

113. 113
     Fedorova OS, Kovshirina AE, Kovshirina YV, et al. Opisthorchis Felineus Infection is a Risk Factor for Cholangiocarcinoma in Western Siberia: A Hospital-based Case-control Study.Clin Infect Dis · 2023 · 76(3):e1392-e1398doi:10.1093/cid/ciac497 · PMID 35723279

     The first human epidemiological evidence for the third liver fluke: 40 histologically confirmed cancer patients matched to 160 controls, with exposure to Opisthorchis felineus carrying an odds ratio of 3.9 (95% CI 1.4-10.8), river fish 5.5 and fishing 3.3. Forty cases in two regions; it is enough to retire the word harmless and not enough to equate this species with the two in Group 1.

114. 114
     Lishai EA, Zaparina OG, Kapushchak YK, et al. Comparative liver transcriptome analysis in hamsters infected with food-borne trematodes Opisthorchis felineus, Opisthorchis viverrini, or Clonorchis sinensis.PLoS Negl Trop Dis · 2024 · 18(12):e0012685doi:10.1371/journal.pntd.0012685 · PMID 39652576

     Cited for the standing of the third liver fluke: it records that O. felineus sits in IARC Group 3 while the other two are Group 1. Its own gloss of Group 3, “noncarcinogenic to humans,” is not what that category means — Group 3 is for agents the evidence cannot classify — and this page does not follow the paper's wording. The study itself compares hamster liver gene expression one and three months after infection with each species, and finds both shared inflammatory pathways and species-specific responses.

115. 115
     Jain S. Exploring the potential role of Opisthorchis felineus infection in cholangiocarcinogenesis.Mol Biochem Parasitol · 2026 · 266:111733doi:10.1016/j.molbiopara.2026.111733 · PMID 41587590

     The current argument for moving the third fluke up the IARC scale: it gathers the animal evidence for DNA repair interference, lipid peroxidation products and oxysterol-like compounds, notes that human carcinogenesis is not established, and concludes that the Group 3 placement should be revised upward. A single-author narrative review from an independent research institute, with no new data and no systematic search, so it is an argument rather than a finding.

116. 116
     Rujirakul R, Ueng-arporn N, Kaewpitoon SJ, Loyd RA, Kaewthani S, Kaewpitoon N. Risk Areas of Liver Flukes in Surin Province of Thailand using Geographic Information System.J Med Assoc Thai · 2015 · 98 Suppl 4:S22-6PMID 26201130

     RETRACTED - cited only to record the retraction. The PubMed record carries the article type Retracted Publication. Its reported figures, including that three factors explained 88.60% of liver fluke infection and that 221.39 square kilometers were heavy-risk area, must not be used.

117. 117
     Sukontason K, Piangjai S, Sukontason K, Chaithong U. Potassium permanganate staining for differentiation the surface morphology of Opisthorchis viverrini, Haplorchis taichui and Phaneropsolus bonnei eggs.Southeast Asian J Trop Med Public Health · 1999 · 30(2):371-4PMID 10774711

     The measured egg dimensions this page uses: O. viverrini 26.34 ± 1.65 by 15.54 ± 0.69 µm, H. taichui 29.03 ± 1.48 by 14.94 ± 0.91 µm, with the two widths not significantly different. It also shows what does separate them — surface texture after staining, melon-like ridges on the liver fluke against light striae on H. taichui. A single laboratory series, and the staining step is not part of routine survey microscopy.

118. 118
     Ditrich O, Giboda M, Scholz T, Beer SA. Comparative morphology of eggs of the Haplorchiinae (Trematoda: Heterophyidae) and some other medically important heterophyid and opisthorchiid flukes.Folia Parasitol · 1992 · 39(2):123-32PMID 1644359

     The source for why egg shape cannot settle which fluke a person has: great variation within each species and great similarity between them, so that “the eggs shape does not seem to be suitable for species identification,” while scanning electron microscopy of the shell surface does distinguish some groups. Descriptive morphology, with no sensitivity figures.