Artemisinin and Cancer: What the Clinical Evidence Actually Shows
An Evidence-Based Review for Clinicians and Informed Clients
Yoon Hang Kim MD
Board-Certified in Preventive Medicine | Integrative & Functional Medicine Physician
About Dr. Kim
Dr. Kim is a board-certified physician with more than 20 years of clinical experience. A University of Arizona/Andrew Weil fellowship-trained integrative medicine specialist, he holds certifications in preventive medicine, medical acupuncture, and integrative and functional medicine. He specializes in low dose naltrexone (LDN), autoimmune conditions, chronic pain, integrative oncology, fibromyalgia, chronic fatigue syndrome, mast cell activation syndrome (MCAS), and mold toxicity. He is the author of eight books — including MCAS: Epidemic in Plain Sight and LDN Primer, both available on Amazon — and more than 25 peer-reviewed articles, and is the founder of the LDN Support Group.
Professional: www.yoonhangkim.com | Clinical: www.directintegrativecare.com
Important: This article is for educational and informational purposes only and does not constitute medical advice. Always consult with a qualified healthcare provider before making changes to your treatment plan.
Abstract
Artemisinin and its derivatives, long established as antimalarial agents, have attracted sustained interest in integrative oncology. This review examines the human clinical trial evidence for their use in cancer, distinguishing laboratory activity from demonstrated clinical benefit. The strongest evidence involves artesunate, a semisynthetic derivative. A single randomized, double-blind, placebo-controlled pilot trial in colorectal cancer (23 randomized, 20 evaluable) reported fewer recurrences with short-course preoperative artesunate (1 versus 6; hazard ratio 0.16, 95% CI 0.02–1.3), but the primary apoptosis endpoint was not met and the recurrence finding was exploratory and statistically nonsignificant. Confirmatory Phase II trials (NeoART, NeoART-V, NeoART-M) have not reported efficacy results. Across other tumor types — non-small cell lung, breast, cervical, and hepatocellular carcinoma — the evidence is confined to early-phase safety and feasibility studies, with no Phase III trial initiated and no regulatory approval for any oncology indication. A critical and under-recognized distinction separates pharmaceutical-grade artesunate from over-the-counter artemisinin supplements, which differ in bioavailability, pharmacokinetics, and quality control. The available evidence positions artesunate as a compound worth continued investigation but insufficient to support use as an established cancer therapy.
Why This Compound Keeps Coming Up
Few repurposed agents generate as much conversation in integrative oncology as artemisinin and its derivatives. Clients arrive with printouts of laboratory studies, links to supplement retailers, and a reasonable question: does this ancient antimalarial compound have a role in cancer care?
The honest answer is more nuanced than either the dismissive “no evidence” or the enthusiastic “proven anticancer herb” that dominate popular discourse. The strongest published randomized human oncology evidence involves artesunate, a semisynthetic derivative, particularly in colorectal and non-small cell lung cancer. Other artemisinin derivatives and Artemisia preparations have entered smaller clinical studies or registered trials, but their evidence is even more limited. And there is a substantial gap between what the research actually tested and what clients are buying at health food stores.
This article reviews the clinical trial evidence across cancer types, identifies where the data is genuinely interesting, where it falls short, and where the disconnect between research-grade compounds and consumer supplements creates real safety concerns. I have drawn on published trial results, registry records, and recent systematic reviews to present what we know as of late 2026.
The Compound: Why Researchers Are Interested
Artemisinin is a sesquiterpene lactone isolated from Artemisia annua (sweet wormwood), the plant behind Youyou Tu’s Nobel Prize-winning antimalarial work. Artesunate is a water-soluble semisynthetic derivative with improved pharmacokinetics — and the compound with the strongest published randomized human cancer evidence. These are not interchangeable substances, and that distinction matters throughout this discussion.
The mechanistic rationale for studying artesunate in oncology centers on its endoperoxide bridge, which reacts with intracellular iron to generate reactive oxygen species. Cancer cells, which tend to accumulate more iron than healthy tissue through upregulated transferrin receptors, may be selectively vulnerable. In laboratory and animal models, artesunate has demonstrated effects on ferroptosis (iron-dependent cell death), Wnt/β-catenin signaling, prostaglandin synthesis, and chemosensitization pathways.
I want to be explicit about what this paragraph means and what it does not. These mechanisms explain why the compound is being studied. They do not constitute evidence that artesunate treats cancer in people. Laboratory tumor-cell killing is a necessary precondition for clinical investigation, not a clinical outcome. Thousands of compounds kill cancer cells in a dish; very few survive the journey to proven clinical benefit.
Colorectal Cancer: The Strongest Human Dataset
The most substantive clinical evidence comes from a single randomized, double-blind, placebo-controlled pilot trial conducted at St George’s, University of London (Krishna et al., EBioMedicine 2015). Individuals with operable Stage I–III colorectal adenocarcinoma received oral artesunate 200 mg daily for 14 days before surgical resection, or a matching placebo.
Of 23 randomized participants, 20 completed per protocol — 9 in the artesunate arm, 11 in the placebo arm. Here is what the trial found, and what it did not.
The primary endpoint was a tissue biomarker: the proportion of tumors with greater than 7% apoptotic cells by TUNEL staining. This occurred in 67% of the artesunate group versus 55% of the placebo group. Because baseline spontaneous apoptosis in the placebo arm was higher than the investigators anticipated, this primary endpoint did not achieve conventional statistical significance. The trial, by its own prespecified standard, was a negative study on its primary outcome.
Bayesian analyses suggested a probable reduction in Ki67 (a proliferation marker), with an estimated 89% to 92% probability under non-informative assumptions. This rose to 97% with an optimistic prior favoring an artesunate effect, but fell to 77% under a skeptical prior centered on no effect. CD31 showed a 79% probability of a treatment effect. Showing this prior-dependence matters: the headline 97% figure is not a robust estimate but the product of a particular assumption. One documented individual with an elevated baseline CEA showed roughly a 75% reduction (from 212 to 56 µg/L) after artesunate alone, prior to surgery.
The recurrence data is what generated the most attention. Over a median follow-up of 42 months, recurrence was reported in 1 artesunate-treated individual versus 6 in the placebo group. The hazard ratio for first recurrence was 0.16, with a 95% confidence interval of 0.02 to 1.3.
That confidence interval deserves a moment. It crosses 1.0, which means the result is not statistically significant by conventional criteria. With only 7 total recurrence events across 20 participants, a single reclassified event could substantially alter the hazard ratio. The magnitude of the point estimate — an 84% relative reduction in the recurrence hazard — is striking, but it emerged from an exploratory analysis in a study too small to detect a survival difference.
The appropriate framing: this is an early signal that warranted larger confirmatory trials, which are now underway. It is not practice-changing evidence, and it should not be cited as proof that artesunate prevents cancer recurrence.
On safety, artesunate was generally tolerated, but two participants developed Grade 3 neutropenia/leukopenia. One also had anemia requiring transfusion; the other had persistent neutropenia requiring G-CSF and experienced a surgical delay. Both were lighter individuals (approximately 50 kg) receiving roughly 4 mg/kg/day. These events prompted later protocols to cap dosing below 4 mg/kg/day and add mid-treatment blood count monitoring.
The Confirmatory Trials: Where We Stand
The pilot’s recurrence signal prompted a series of Phase II trials designed to test it in statistically powered cohorts. None have yet published efficacy results.
The UK NeoART trial (NCT02633098) was designed to enroll 200 participants with operable Stage II/III colorectal cancer, using the same 14-day preoperative artesunate 200 mg regimen, with recurrence-free survival as its primary endpoint. Recruitment was disrupted during the COVID-19 pandemic. Its ClinicalTrials.gov record has not been updated since October 2022 and is now classified as “unknown status”; no efficacy results have been posted.
NeoART-V (NCT03093129), a parallel study in Vietnam, has not reported efficacy results; its registry record is stale and listed as unknown, though recent reviews describe the study as suspended.
Most recently, NeoART-M (NCT07095309) has been organized in Malaysia, again using the same protocol: oral artesunate 200 mg daily for 14 days before surgery in Stage II/III CRC, with recurrence outcomes followed long-term. As of its October 2024 protocol document, estimated completion extends into the early 2030s.
The pattern here is important to recognize. A provocative pilot generated confirmatory trials, but recruitment challenges, the pandemic, and the logistics of international multi-site oncology research have delayed the answer. The question is genuinely being tested — but it has not been answered. Anyone who tells you artesunate has been proven to reduce colorectal cancer recurrence is outrunning the evidence.
Beyond Colorectal Cancer: An Evidence Map
Outside of CRC, human clinical data on artesunate in oncology is sparse and early-phase. Here is what exists, organized by the level of evidence available rather than the excitement of the mechanism.
Non-small cell lung cancer (NSCLC). A randomized controlled trial in China (Zhang et al., 2008) enrolled 120 participants with advanced NSCLC, comparing vinorelbine/cisplatin chemotherapy with or without intravenous artesunate (120 mg daily for 8 days). No significant differences were observed in overall survival, one-year survival, or quality of life. The artesunate group did show a higher disease control rate (88.2% versus 72.7%) and a modestly longer time to progression (24 versus 20 weeks). This was a different compound formulation (IV, not oral), a different dose and schedule, and a different cancer from the CRC pilot. It should be read on its own terms: a signal of possible disease-control benefit with combination chemotherapy, without a survival advantage.
Metastatic breast cancer. A Phase I study (ARTIC M33/2, von Hagens et al., 2017) evaluated oral artesunate 100 to 200 mg daily for four weeks alongside guideline-based cancer therapy. This was primarily a safety and dose-finding study. A compassionate-use follow-up (2019) described long-term tolerability in individuals continuing oral artesunate. Neither study was designed to demonstrate antitumor efficacy.
Cervical cancer. A proof-of-concept trial evaluated intravaginal artesunate for cervical intraepithelial neoplasia (CIN2/3), representing the earliest stage of clinical investigation. A Phase I dose-escalation study of oral artesunate in metastatic cervical carcinoma was completed in Kenya. Some reviewers have noted that cervical carcinoma may respond relatively better to artesunate than other solid tumors, but this remains a preliminary observation.
Hepatocellular carcinoma (HCC). A Phase I dose-escalation study at University Hospital Ghent evaluated oral artesunate in advanced HCC. The study was terminated due to slow recruitment and expiration of the study drug — partly because checkpoint inhibitor trials opened for the same population. No efficacy results are available.
Other solid tumors. A Phase I study of intravenous artesunate in individuals with advanced solid tumors (Deeken et al., 2018) established safety and pharmacokinetic data. Individual case reports exist in the literature, but case reports do not establish treatment efficacy.
The picture across all tumor types is consistent: early-phase safety and feasibility work, with no Phase III trial initiated in any cancer and no regulatory approval anywhere for an oncology indication.
What the Trials Did Not Test: The Supplement Problem
This is the section I consider most clinically important, because it addresses the gap between what was studied and what clients are actually taking.
The clinical trials described above used pharmaceutical-grade artesunate — a semisynthetic, water-soluble derivative of artemisinin with defined pharmacokinetic properties, manufactured under pharmaceutical quality standards, and administered at specific doses under medical supervision with blood count monitoring.
What clients buy at supplement retailers is typically one of these: dried Artemisia annua herb or tea, a crude herbal extract of Artemisia annua, or an isolated artemisinin compound in capsule form. None of these is artesunate. None was tested in the trials cited above. And the differences are not trivial.
First, bioavailability. Oral artemisinin has poor and variable absorption compared to artesunate. The amount of active compound reaching the bloodstream from an herbal capsule is substantially lower and less predictable than from pharmaceutical artesunate.
Second, pharmacokinetics differ substantially among artemisinin compounds. Artemisinin itself demonstrates pronounced, reversible auto-induction of its own metabolism during repeated oral dosing. In healthy volunteers, systemic exposure falls markedly over several consecutive days — artemisinin AUC has been reported to drop to roughly 20 to 24% of day-1 values by day 7 (Ashton et al., 1998), mediated by induction of cytochrome P450 enzymes (notably CYP2B6 and CYP2C19). Similar time-dependent pharmacokinetics have been reported for some related compounds such as artemether, but the magnitude and clinical relevance cannot simply be assumed to be identical for artesunate, where the evidence is less convincing. This makes extrapolation from chronic OTC artemisinin supplementation to the pharmaceutical artesunate used in trials particularly problematic — a client taking an artemisinin supplement daily for months is not reliably reproducing any studied exposure.
Third, dose standardization. Consumer supplements are not held to the same manufacturing standards as pharmaceutical compounds. Batch-to-batch variation in active compound content, the presence of other plant constituents, and the absence of quality testing mean that what is on the label may not reflect what is in the bottle.
The practical consequence: a client reading about artesunate’s recurrence signal in colorectal cancer and then purchasing an artemisinin supplement online is not replicating the trial. They are taking a different compound, at an uncertain dose, with declining exposure over time, without the blood count monitoring that caught the neutropenia events in the pilot study.
This is not a reason to dismiss interest in the compound. It is a reason to be precise about what was studied, what was not, and what the implications are for self-directed supplementation.
Calibrating the Evidence: Where Artesunate Sits
It helps to locate artesunate’s oncology evidence relative to other integrative agents that clinicians discuss. This is not a head-to-head comparison — these compounds address different clinical questions in different cancers — but a calibration of what different levels of evidence look like in practice.
For context on what mature adjunctive evidence looks like: an NCI-reviewed meta-analysis (Oba et al., 2007) of eight randomized trials involving 8,009 individuals after gastric cancer surgery found that adding standardized PSK (a polysaccharide from Trametes versicolor, the turkey tail mushroom) to chemotherapy improved overall survival with a hazard ratio of 0.88 (95% CI 0.79–0.98). That is a modest but statistically significant effect across thousands of individuals in a specific cancer, with a specific standardized product, added to specific chemotherapy regimens. The FDA has not approved PSK as a cancer treatment, but the evidence tier is qualitatively different from a 20-person pilot.
For context on what mature supportive-care evidence looks like: a randomized trial (Barton et al., 2013) of 364 cancer participants found clinically meaningful improvement in cancer-related fatigue at eight weeks with American ginseng (Panax quinquefolius) 2,000 mg daily. That trial addressed symptom management — a legitimate clinical goal — not tumor control.
Artesunate’s colorectal cancer pilot, with 20 evaluable individuals and a secondary finding of reduced recurrence, sits well below both of those evidence thresholds. It is a hypothesis-generating signal from a compound with a plausible mechanism and an interesting pharmacological profile. It is not in the same evidence tier as PSK’s adjunctive survival data or American ginseng’s fatigue data — and clinicians should be careful not to present it as though it were.
Clinical Takeaway
When clients ask me about artemisinin and cancer, my response is structured around three honest distinctions.
First, what was studied matters. The clinical research used pharmaceutical-grade artesunate, not artemisinin supplements and not Artemisia annua tea. These are pharmacologically different preparations with different absorption, metabolism, and exposure profiles. Extrapolating from one to another is not clinically appropriate.
Second, the level of evidence matters. Artesunate has a genuinely interesting pilot signal in colorectal cancer — a recurrence reduction that, if confirmed, would be clinically meaningful. But it has not been confirmed. The confirmatory trials have not reported results. No Phase III trial has been initiated in any cancer type. No regulatory body has approved artesunate for any oncology indication. Presenting this compound as an established cancer treatment misrepresents the evidence.
Third, the clinical goal matters. Are we discussing a potential adjunct to conventional treatment under medical supervision and monitoring? Or are we discussing a client self-treating with an unregulated supplement as a substitute for surgery, chemotherapy, or immunotherapy? The first is a legitimate area of clinical investigation. The second is a safety concern.
The research on artesunate deserves to continue. The NeoART series of trials may eventually provide the statistical power to determine whether the pilot’s recurrence signal is real. In the meantime, clients deserve clinicians who can hold two ideas simultaneously: that a compound can be genuinely worth investigating and genuinely insufficient as a treatment recommendation.
That is not a hedge. It is what intellectual honesty looks like when the evidence is still being gathered.
Conclusion
The evidence for artemisinin compounds in cancer occupies a genuinely intermediate position that resists both easy dismissal and premature enthusiasm. Artesunate carries one provocative randomized pilot signal in colorectal cancer and a plausible mechanism, which together justify the confirmatory trials now underway. But a signal is not a conclusion. The pivotal trial was small, its primary endpoint was negative, and its recurrence finding was exploratory and statistically nonsignificant. No larger trial has confirmed it, and no artemisinin compound is approved for any cancer indication anywhere in the world.
Until that changes, the responsible clinical posture is calibrated interest: following the NeoART results closely, keeping pharmaceutical artesunate and retail artemisinin products firmly separate in our thinking, and declining to let an intriguing early signal stand in for established cancer therapy. If confirmatory data arrive, this assessment will warrant revision — and I will be glad to revise it.
Yoon Hang Kim MD is the author of MCAS: Epidemic in Plain Sight, LDN Primer, and Integrative Oncology: Evidence-Based Strategies to Support Cancer Treatment, all available on Amazon. He is the founder of the LDN Support Group.
Work with Dr. Kim
If you are navigating a cancer diagnosis and want evidence-based integrative guidance — or if you have questions about supplements like artemisinin in the context of your treatment plan — Dr. Kim offers integrative and functional medicine consultations through his telemedicine practice, including consultation, IV nutrition, functional medicine testing, and specialty acupuncture.
Learn more and schedule a consultation at www.directintegrativecare.com.
References
1. Krishna S, Ganapathi S, Ster IC, et al. A randomised, double blind, placebo-controlled pilot study of oral artesunate therapy for colorectal cancer. EBioMedicine. 2015;2(1):82-90. doi:10.1016/j.ebiom.2014.11.010
2. Zhang ZY, Yu SQ, Miao LY, et al. Artesunate combined with vinorelbine plus cisplatin in treatment of advanced non-small cell lung cancer: a randomized controlled trial. Zhong Xi Yi Jie He Xue Bao. 2008;6(2):134-138. doi:10.3736/jcim20080206
3. von Hagens C, Walter-Sack I, Goeckenjan M, et al. Prospective open uncontrolled phase I study to define a well-tolerated dose of oral artesunate as add-on therapy in patients with metastatic breast cancer (ARTIC M33/2). Breast Cancer Res Treat. 2017;164(2):359-369. doi:10.1007/s10549-017-4261-1
4. von Hagens C, Walter-Sack I, Goeckenjan M, et al. Long-term add-on therapy (compassionate use) with oral artesunate in patients with metastatic breast cancer. Phytomedicine. 2019;54:140-148. doi:10.1016/j.phymed.2018.09.178
5. Deeken JF, Wang H, Hartley M, et al. A phase I study of intravenous artesunate in patients with advanced solid tumor malignancies. Cancer Chemother Pharmacol. 2018;81(3):587-596. doi:10.1007/s00280-018-3533-8
6. Trimble CL, Levinson K, Maldonado L, et al. A first-in-human proof-of-concept trial of intravaginal artesunate to treat CIN2/3. Gynecol Oncol. 2020;157(1):188-194. doi:10.1016/j.ygyno.2019.12.035
7. Oba K, Teramukai S, Kobayashi M, Matsui T, Kodera Y, Sakamoto J. Efficacy of adjuvant immunochemotherapy with polysaccharide K for patients with curative resections of gastric cancer. Cancer Immunol Immunother. 2007;56(6):905-911. doi:10.1007/s00262-006-0248-1
8. Barton DL, Liu H, Dakhil SR, et al. Wisconsin ginseng to improve cancer-related fatigue: a randomized, double-blind trial, N07C2. J Natl Cancer Inst. 2013;105(16):1230-1238. doi:10.1093/jnci/djt181
9. CAM-Cancer. Artemisia annua [monograph]. Updated 2026. cam-cancer.org/artemisia-annua
10. NeoART-M Phase II Trial Protocol, Version 6.0. October 2024. ClinicalTrials.gov: NCT07095309.
11. NeoART Trial. ClinicalTrials.gov: NCT02633098. Unknown status (last updated October 2022).
12. NeoART-V Trial. ClinicalTrials.gov: NCT03093129. Registry record stale; reviews describe as suspended.
13. Ashton M, Hai TN, Sy ND, et al. Artemisinin pharmacokinetics is time-dependent during repeated oral administration in healthy male adults. Drug Metab Dispos. 1998;26(1):25-27.
