A couple of readers have asked about a purported cancer therapy agent called artemisinin. This is a compound derived from a plant and used for a variety of purposes in the system known as Traditional Chinese Medicine. It is sometimes recommended outside of this system as an herbal therapy for cancer, with the usual claims of dramatic benefits and virtually no risk of side effects. As we’ll see, the truth is more complex and more interesting than such simplistic claims.
What Is It?
A number of compounds have been identified in the plan known most commonly as Sweet Wormwood. Artemisinin is the most widely studied of these and the one most often claimed to have medicinal benefits. The usual forms in which the drug is given is as a dried herb or infusion (steeping the plant in water like tea).
It has been claimed to have benefits in treating cancer and fungal or parasitic infections in pets. It is also widely used, in conjunction with other drugs, in treatment of malaria in humans.
Does It Work?
There is no question that artemisinin has been useful in the treatment of malaria. Like any effective medicine, of course, it has been shown to have side effects (relatively low rates of nausea and vomiting, rare allergic reactions and possibly liver damage), and there is now evidence that malaria is developing resistance to artemisinin-based therapy.
The evidence for use in cancer therapy is not nearly as convincing (see also this review). There are a number of in vitro and lab animal studies showing potentially useful effects on cancer cells. However, this has been true for most cancer therapies that have gone on to be failures in actual clinical trials, so all it establishes is the possibility the drug might be useful, not that it actually is.
The next level of evidence is the case report, which is simply a formalized, published version of the anecdote. There are some case reports showing improvement in human cancer patients with use of artemisinin and related compounds (e.g. 1, 2). However, as with all anecdotes, these reports are subject to enormous risk of bias and confounding, and it is not appropriate to draw conclusions about the safety and efficacy of artemisinin from them. At most, they suggest the possibility that further, controlled testing is appropriate.
There isn’t much in the way of such evidence in humans. One clinical trial combining the compound with chemotherapy for lung cancer patients found no difference in terms of short, medium, or long-term survival. Some difference was seen in the rate of disease progression, but the reality and clinical relevance of this is doubtful, and the source is pretty questionable since it is a Chinese alternative medicine journal, and such journals appear to never publish negative results.
There is, similarly, virtually no clinical trial literature in veterinary species evaluating artemisinin as a cancer therapy. A clinical trial investigating the related compound artesunate as a therapy for lymphoma in dogs has recently been published:
J.N. Bryan, D. Tate, S.M. Bechtel, K. Choy, L. Donnelly, K. Fitzpatrick, B.K. Flesner, B. Fowler, B. Gallis, C.J. Henry, C. Herrera, M. Jabbes, K.A. Selting, T. Sasaki. Randomized, blinded, placebo-controlled trial of artesunate with doxorubicin for B-cell lymphoma of dogs. J Vet Int Med. 2014;28(4):1362.
Artemisinin has described anti-cancer properties. Artemisinin can down-regulate VEGF, HIF-1alpha, and survivin in cancer cells, causing apoptotic death. Artesunate, a water-soluble derivative, has better oral absorption in dogs than artemisinin. The hypothesis was that oral artesunate is tolerable with doxorubicin and results in longer disease-free interval (DFI) than doxorubicin and placebo in dogs with stage 3a or 4a B-cell lymphoma.
Twenty-one qualifying dogs were randomized (blinded) to receive artesunate 20 mg/kg or equivalent placebo. Dogs received study drug alone for 72 hours, then received 30 mg/kg doxorubicin q.3 weeks for three doses with study drug. Thereafter, lymph node measurement, CBC, chemistry profile, and urinalysis were monitored monthly until disease progression. Differences and DFI were compared by Mann-Whitney and Kaplan-Meier logrank tests. Significance was set at P ? 0.05.
Artesunate alone did not cause clinical remission (P = 0.1347). Artesunate induced no biochemical toxicosis. One dog experienced progressive disease following doxorubicin; all others experienced a complete response. Artesunate caused no greater neutropenia or thrombocytopenia following doxorubicin (P = 0.6234 and P = 0.2311). DFI was not different between groups (70d for placebo and 122d for artesunate, P = 0.967). Dogs receiving artesunate had higher nucleated red blood cell counts (median 272nRBC/100WBC artesunate and 9nRBC/ 100WBC placebo, P = 0.0142). Red cells displayed poikilocytosis and basophilic stippling in artesunate-treated dogs.
Artesunate did not cause remission as a single agent, nor improved DFI in dogs treated with doxorubicin. 20 mg/kg artesunate caused evidence of heme stress in treated dogs. Artesunate was otherwise well tolerated with doxorubicin. Artesunate cannotbe recommended as a therapy for B-cell lymphoma in dogs.
This trial, like that reported in humans, found no evidence that artesunate improved survival used in conjunction with chemotherapy, and it did not appear to have any beneficial effects used alone.
Another uncontrolled clinical trial in dogs was identified some possible toxicity as well as some possible benefits, but was the heterogeneity of patients and treatments make any firm conclusions impossible.
So far, the clinical trial evidence does not strongly support the preliminary in vitro and lab animal studies or case reports that suggested these compounds might have benefits in cancer patients. This is not unusual, since the vast majority of drugs fail to become useful treatments in actual patients even when they initially look promising in pre-clinical studies. More research may help clarify the value, if any, of artemisinin in cancer therapy.
Is It Safe?
A number of potential risks to artemisinin have been reported in humans and other species. Generally, side effects have been mild and rare in humans, with low rates of vomiting and other GI symptoms reported. There have been concerns about neurologic toxicity, including hearing damage, but it is not yet clear how significant this risk really is. In the context of malaria treatment, where there are clear benefits, the risks appear to be small enough in most cases (other than pregnant women and, possibly, young children) to justify use. However, the risks in cancer patients have not been established.
In dogs, there have not been safety studies in cancer patients either, however there is stronger evidence of risk with artemisinin use. One study of various forms and routes of administration found anemia, neurotoxicity, and changes in mitochondria. Other studies have also found significant toxicity at high doses (see also this study). However, other studies using different forms and dosages have not reported adverse effects.
There is clear clinical trial evidence that artemisinin is useful in the management of malaria in humans, though there are some concerns about it losing its efficacy. Side effects are reported, but they are usually mild and uncommon.
There is some pre-clinical evidence (in vitro studies and lab animal studies) suggesting artemisinin and related compounds might have some benefit in cancer treatment. This is supported by a few uncontrolled case reports in humans. However, the limited clinical trial evidence does not generally support any real benefit for actual patients, human or canine. More research is needed to clarify whether there are circumstances in which the drug might be helpful, but it is by no means a reliable, proven cancer therapy.
Pre-clinical trials in dogs have found some potentially significant risks, which seem to be greater than those seen in humans. Some of these, such as changes in red blood cell production, have also been seen in the limited clinical trials done in dogs, but others have not.
Overall, there is enough evidence to support further controlled clinical research but not to use the treatment in patients outside the context of appropriate clinical trials designed to assess safety and efficacy.