Ivermectin and Cancer in 2026: What the Evidence Shows, What It Doesn’t, and What Prescribers Are Being Asked

Few drugs have travelled a stranger path in the last five years than ivermectin. An antiparasitic that earned its discoverers a Nobel Prize, it became a pandemic flashpoint, and since early 2025 it has been the most-searched cancer question many prescribers hear in clinic. Patients arrive having read that it cures tumors; clinicians arrive having read that it does nothing. Both are wrong, and the actual picture is more interesting than either. This is a plain summary of where the evidence stands as of September 2026, what has changed in the law, and how it intersects with the compounding pharmacies on PEPTPlus.

This article is educational and is not medical advice. Ivermectin is not approved by the FDA to treat any cancer, and nothing here should be read as a recommendation to use it for one. Treatment decisions belong to a patient and their licensed clinician.

Why everyone is asking now

The interest is real and measurable. An analysis presented at ESMO 2025 traced internet search trends for “ivermectin cancer” from 2004 onward and found the curve began in April 2021 and peaked in January 2025, with the United States leading. That January peak has a specific cause: a celebrity described ivermectin and fenbendazole as cancer cures on a widely heard podcast. A UCLA Health study of national prescribing records found that ivermectin prescribing roughly doubled in the following months compared with the prior year, and among people with a cancer diagnosis it ran more than 2.5 times higher. A follow-on EHR analysis at ASCO 2026 described the same pattern: still uncommon in absolute terms, but a rapid, largely patient-driven rise.

So the question is not going away. What a prescriber owes a patient who asks is an accurate account of the evidence, and that is what the next three sections try to give.

The preclinical case is genuinely interesting

The laboratory literature is why serious oncologists have not simply dismissed the idea. In cell lines and animal models, ivermectin has shown activity against a wide range of tumor types through several mechanisms at once: interference with PAK1 signaling, the Wnt/β-catenin pathway, and mitochondrial function, plus effects on cancer stem cells and, in some models, on the tumor immune environment. A 2023 review in the International Journal of Molecular Sciences catalogues those results across more than twenty cancer types. That is a legitimate scientific rationale, and it is the reason there are human trials at all.

It is also the point where the internet version of the story usually stops, and it matters why. Preclinical activity is common. Most compounds that kill cancer cells in a dish or shrink tumors in mice do not help people, because human doses, pharmacokinetics, and tumor biology are different. The concentrations that produce anticancer effects in many ivermectin studies are higher than what standard antiparasitic dosing achieves in blood. None of that makes the preclinical work wrong; it makes it a hypothesis.

What the human data actually shows

The first real trial: ivermectin plus immunotherapy in triple-negative breast cancer

The most important human evidence comes from a Phase I/II trial (NCT05318469) led by Peter P. Lee, MD, of City of Hope, testing oral ivermectin combined with a PD-1 checkpoint inhibitor (balstilimab, and later pembrolizumab) in metastatic triple-negative breast cancer. The rationale was that ivermectin drove T-cell infiltration into tumors in mouse models, turning “cold” tumors “hot” and potentially making immunotherapy work where it otherwise would not.

The preliminary results presented at ASCO 2025 are worth stating exactly, because they are quoted in both directions. Among eight evaluable, heavily pretreated patients, one had a partial response, one had stable disease, and six had progressive disease. Median progression-free survival was 2.5 months, and the four-month clinical benefit rate was 37.5 percent. The combination was safe and well tolerated. The investigators concluded that the signal warranted continued study, and the trial is ongoing with an expansion cohort.

Read plainly: this is a small early-phase safety study with one response, in a disease where responses are hard to come by. It is not a null result, and it is not a cure. It is the kind of result that justifies the next trial, which exists: ICONIC (NCT07487805), a randomized Phase II study of ivermectin at different doses combined with immune checkpoint inhibitors in solid tumors, planned to enroll about 80 patients with a mid-2026 start. That trial, not any anecdote, is where the answer will come from.

The observational cohort, and why the journal flagged it

The study most often shared as proof is a prospective observational cohort published in Anticancer Research in June 2026: 197 patients with cancer prescribed compounded ivermectin (25 mg) and mebendazole (250 mg) capsules off-label through a U.S. telemedicine platform. Among the 122 who completed follow-up, the authors reported a self-reported clinical benefit rate of 84 percent and good tolerability, and they described the findings as hypothesis-generating.

Two things a prescriber should know before citing it. First, the design cannot establish that the drugs did anything: no control group, outcomes largely self-reported, and 38 percent of participants lost to follow-up, which in an observational cohort tends to inflate apparent benefit. Second, the journal’s editorial board has since issued a formal Expression of Concern citing serious questions raised about the verifiability and statistical reliability of the dataset and its ethical oversight. Until those are resolved, the paper should be treated as unverified.

Mebendazole and fenbendazole are not the same story

Because the three drugs travel together online, it is worth separating them. Mebendazole, a human antiparasitic, has the longest oncology history of the three, including a positive randomized trial in colorectal cancer and a negative one in glioblastoma, and it remains under study in brain tumors (a 2023 review summarizes the field). Fenbendazole is a veterinary drug with no approved human use and no human trial data at all; its popularity rests on anecdotes. Grouping it with ivermectin borrows credibility it has not earned.

What the professional bodies say

In May 2026, ASCO issued a Clinical Notice recommending against ivermectin and fenbendazole for cancer treatment, whether alone or as an adjunct, outside a well-designed clinical trial. Its reasoning is the same as the summary above: no robust peer-reviewed evidence of benefit in any human malignancy, real potential for toxicity and drug interactions at the doses being used, and a documented wave of self-medication driven by social media. ASCO paired the notice with guidance urging oncologists to raise the topic proactively rather than wait for patients to disclose it, which many do not.

The FDA’s position is unchanged: ivermectin (Stromectol) is approved for two parasitic infections, strongyloidiasis and onchocerciasis, and for nothing else. Any cancer use is off-label.

The law moved faster than the science

While the evidence has stayed early-stage, access has changed materially. Tennessee made ivermectin available without a prescription in 2022, and in 2025 Arkansas, Idaho, and Louisiana followed, with Texas’s law taking effect in December 2025 and more bills pending. These laws generally declare ivermectin suitable for over-the-counter sale but include no dosing guidance, and they do not obligate any pharmacy to stock it; several chains have declined, leaving independents to decide store by store.

The practical effect for clinicians is that a growing share of patients can obtain human ivermectin with no clinician involved, and a share of the rest are obtaining veterinary paste, which is neither dosed nor purified for people. Both are reasons to ask the question directly at intake.

Where PEPTPlus and its pharmacies fit

Many of the compounding pharmacies on PEPTPlus prepare ivermectin, and prescribers ask us how that squares with everything above. The answer is the same principle that runs through the platform: the prescribing decision is the clinician’s alone, and our job is to make sure that when a decision is made, the sourcing and the record are beyond question.

• We do not promote off-label use. PEPTPlus takes no position on whether ivermectin is appropriate for any patient or indication, and nothing in the platform nudges a prescriber toward it. There is no per-prescription revenue on either side, so nobody benefits from a script being written.
• Every fill is a human-grade, pharmacy-dispensed compound with a batch COA. Ivermectin compounded under 503A is prepared from pharmaceutical-grade material by a state-licensed pharmacy, and the certificate of analysis for the batch is attached to the order. That is the opposite of a tube of horse paste, and it is the difference that matters if a patient is going to take the drug anyway.
• Routing is by licensure, not by demand. A prescription goes only to a pharmacy licensed to dispense into the patient’s state, matched on verified quality, price, and delivery, never on which pharmacy would most like the order.
• The rationale lives with the prescription. A prescriber who chooses ivermectin off-label can document why, in the record, at the time. If a regulator, a payer, or a colleague asks later, the answer is one click away rather than a reconstruction.

If you are an oncologist, the honest counsel to a patient who asks is: the science is early, the one real trial is small and ongoing, the widely shared cohort is under a cloud, and the professional consensus is trial-only. If you are a prescriber who has nonetheless decided, with a patient, that a supervised off-label course is reasonable in their circumstances, then the least you owe them is a verified product, a licensed pharmacy, and a written reason. That part we can help with. You can start prescribing on PEPTPlus, or read how verified sourcing works first.