Fenbendazole

What it actually is. Fenbendazole is a benzimidazole carbamate anthelmintic licensed for dogs, cattle, horses and other animals. Its relatives albendazole and mebendazole are licensed for human parasitic infections; fenbendazole is not approved for humans anywhere [1].

Where the cancer claim came from. In 2019 an American patient with small-cell lung cancer publicised his recovery and attributed it to fenbendazole taken alongside conventional treatment. The claim spread rapidly online, and self-administration followed - notably in South Korea, where the phenomenon was written up as "fenbendazole fever" [8]. The published commentary at the time was cautious, asking whether anthelmintics could be anticancer drugs and concluding the question was open [9].

What the preclinical evidence shows. It is real, and it is preclinical. Fenbendazole has antitumour activity against cervical cancer cells and xenografts, acting on both cancer cells and cancer stem cells [2]. Transcriptome analysis with in vitro and in vivo work supports activity in ovarian cancer [3], and PLGA nanoparticle formulation improves it by addressing the compound's poor solubility [10]. A 2024 review argued that fenbendazole's low cost, safety profile in animals and anti-proliferative activity make it the benzimidazole of choice to investigate - and stated explicitly that clinical trials are crucial before repurposing, because its human pharmacokinetics are unknown [1].

What the human evidence shows. There is no clinical trial. None. The publication most often cited as human evidence - a 2025 case series of three self-administering patients in Case Reports in Oncology - has been retracted [7]. A retracted paper is not weak evidence; it is withdrawn from the record.

What is documented in humans is harm. A histologically confirmed case of severe hepatocellular drug-induced liver injury from self-administered fenbendazole, normalising three months after stopping [4]. A cancer patient who took 2 g daily and developed cholestasis with lobular and portal inflammation; the case report notes fenbendazole is reported to cause cholangitis and liver failure in humans [5]. And a patient on nivolumab/relatlimab who developed severe hepatocellular injury after increasing her self-administered fenbendazole dose - a structured causality assessment scored fenbendazole as probable, and immunotherapy was successfully reintroduced afterwards with no recurrence [6].

That last case is the most consequential. Liver injury in a patient on a checkpoint inhibitor is ordinarily attributed to the immunotherapy, which means stopping a working cancer treatment. Here it was the dewormer.

Benzimidazoles bind beta-tubulin and prevent microtubule polymerisation. In parasitic worms this collapses the cytoskeleton and starves the organism; the selectivity comes from a higher affinity for nematode tubulin than for mammalian tubulin [1].

The anticancer hypothesis rests on that residual mammalian activity: microtubule disruption is a well-established anticancer mechanism - vinca alkaloids and taxanes work that way - and rapidly dividing tumour cells are more vulnerable to it than most normal tissue. Preclinical work reports additional effects on cancer stem cells and on transcriptional programmes [2][3].

The gap between that and a treatment is the usual one, and here it is unusually wide. Fenbendazole is barely soluble and poorly absorbed, so the concentrations that kill cells in a dish may be unreachable in a human tumour at any tolerable oral dose - which is exactly why researchers have turned to nanoparticle formulations [10]. Nobody has measured what plasma or tissue concentrations human self-dosing actually produces [1]. Without that, the preclinical potency figures cannot be translated into anything.

Sold as veterinary granules, paste and oral suspension, formulated and dosed for animals. There is no human-use product, no human dosage form and no pharmaceutical quality standard for human consumption.

The poor aqueous solubility that limits absorption is the formulation problem the research literature is trying to solve, so far only in animals [10].

In humans the documented events are liver injuries. Three published case reports describe severe drug-induced liver injury in people taking fenbendazole for cancer: histologically confirmed hepatocellular injury that normalised three months after cessation [4]; cholestasis with lobular and portal inflammation in a patient taking 2 g daily [5]; and severe hepatocellular injury in a patient on combination immunotherapy [6]. The Norwegian case report notes fenbendazole is reported to cause cholangitis and liver failure in humans [5].

The immunotherapy confound is a specific, serious hazard. Checkpoint-inhibitor hepatitis and fenbendazole-induced liver injury look alike. Misattributing the second to the first means stopping effective cancer treatment and starting steroids for the wrong reason. The 2026 case report exists precisely to make that point, and concludes that proactive patient counselling is needed because of the significant liver risks [6].

The "it's safe in animals" argument does not transfer. A wide margin in dogs at deworming doses for a few days says nothing about a human taking grams daily for months. Human pharmacokinetics have never been characterised [1].

On the retracted case series. The three-patient series in Case Reports in Oncology reporting benefit from self-administration has been retracted [7]. It should not be cited as evidence, and anyone who encountered the fenbendazole claim through it is working from a withdrawn source.

The documented interaction concern is with cancer immunotherapy, and it is one of diagnosis rather than pharmacology: fenbendazole-induced liver injury is readily mistaken for checkpoint-inhibitor hepatitis in a patient on nivolumab or similar agents, with the consequence that the immunotherapy is stopped unnecessarily [6].

Benzimidazoles more generally are substrates and modulators of hepatic metabolism, and hepatotoxicity risk would be expected to compound with other hepatotoxic drugs. No formal interaction studies exist for human use, because there is no human use to study.

Anyone taking this alongside cancer treatment should tell their oncologist. That is the single most useful thing on this page.