Green Templeton College – anamed – RITAM workshop on

Artemisia annua for the treatment of malaria

Saturday, March 13^th 2010

Barclay Room, Green Templeton College, Oxford

Report

Organisers: Dr Merlin Willcox (Department of Primary Health Care), Dr Keith Lindsey (anamed),
Dr Elisabeth Hsu (Institute of Social and Cultural Anthropology), Chair: Dr Simon Challand (anamed)

Aim: To discuss the way forward for Artemisia annua as a herbal antimalarial, from the perspectives of researchers and of NGO / public health programmes.

Objectives:

• To present a synthesis of research – what evidence we do and don’t have on efficacy and safety, and priorities for future research.
• To discuss existing programmes on cultivation and use of Artemisia annua and how these could be improved.
• To discuss how Artemisia annua could fit in to public health programmes for the control and treatment of malaria.
• To discuss the issue of resistance to artemisinin derivatives, and whether this is more/less likely if Artemisia annua is used on a large scale.
• To discuss priorities for future research.

• Introduction

This multidisciplinary workshop was attended by 32 people from a variety of backgrounds: anthropologists, agronomists, biochemists, botanists, herbalists, primary care clinicians, pharmacologists, public health specialists, and NGO workers and trustees. The organizers are grateful to Green Templeton College for providing the room and to the Department of Primary Health Care for sponsoring catering.

Anamed and the promotion of herbal Artemisia annua (Dr Keith Lindsey)

Anamed (Action for Natural Medicine) (www.anamed.net) is an NGO based in Germany which was founded by the pharmacist Hans-Martin Hirt after being in charge of a hospital pharmacy in the Congo where he had to deal with frequent stock-outs of medications. Anamed places a strong emphasis on valuing and using local knowledge, skills and resources, drawing together local unwritten knowledge in Africa and scientific research. It aims to combine the advantages of modern medicine (scientific basis, reliable preparation, hygienic packaging, labeling, clear dosages, expiry dates) with the advantages of traditional medicine (using local resources, accords with local culture, and almost always available).

In this context it negotiated the purchase of high-yielding Artemisia annua seeds from a commercial company for humanitarian purposes, on the condition that the company should remain anonymous. Anamed started planting Artemisia annua in the Democratic Republic of the Congo (DRC) in 1997 and using a herbal tea to treat malaria. The dosage used was that recommended in the Chinese Pharmacopoeia (5-9g of dried leaves in 1litre of water per day for 7 days). Anamed now has partners in 75 countries growing Artemisia annua and using the tea to treat malaria. This also results in economic benefits to communities who grow Artemisia and sell the tea as a treatment for malaria. Anamed collects anecdotal evidence and people claim this is the best treatment and keeps patients malaria free for longer. Anamed doesn't recommend Artemisia as a malaria prophylaxis but does have reports from groups who claim to use it in this way with good effect. Anamed believes that the use of the whole plant extract with possibly 10 active antimalarial compounds (according to the ethnobotanist Dr James Duke) contributes to its apparent effectiveness.

Evaluation and pharmacovigilance of anamed projects in Uganda and Kenya (Shelly Burton and Merlin Willcox)

Pharmacovigilance is the detection, assessment, understanding and prevention of adverse effects and other drug-related problems. There are very few pharmacovigilance systems for herbal or traditional medicines particularly in Africa. Artemisia annua is widely recognized as safe, except for concerns about its safety in early pregnancy. Anamed does not recommend use of Artemisia in the first trimester of pregnancy but many women may use it without knowing they are pregnant.

In July 2009 Shelly Burton interviewed ten natural medicine practitioners (Anamed partners) in three regions of Western Uganda about any adverse effects observed, particularly in pregnant women. One doctor had followed a cohort of 20 women who had taken Artemisia annua tea in the first trimester. They had all delivered normally except for one case of low birth weight. In general other practitioners reported that it is difficult to follow up a patient for the duration of their pregnancy. It might be possible to do this in the future with mobile phones.

In November 2009 Merlin Willcox visited eight Anamed partners in Western Kenya, and another nine partners completed a pilot questionnaire on the pharmacovigilance and general evaluation of Artemisia annua cultivation and use. The plants were mainly distributed through the network of churches and REAP (Rural Extension with Africa’s Poor, www.reap-eastafrica.org ). This is probably the best infrastructure available in this area, and in one case plants had been distributed to over 200 people by one pastor. Furthermore Artemisia annua fits neatly into the context of other development projects promoting self-reliance. However, although it may not be relevant in this area, thought needs to be given to how to reach non-churchgoers and those of other faiths.

Cultivation had been adversely affected by the weather in two cases – one partner in the highlands had lost seedlings in hailstorms, while another partner in the lowlands has not been able to grow good sized plants because of drought. Some partners were uncertain whether they could use self-sown seedlings from the plants or whether they should only use cuttings as recommended by Anamed. In practice people were using the seedlings also, although their artemisinin content is probably lower than in the parent plants.

Eleven of 17 respondents said that they treated children with herbal Artemisia annua, but often with an inadequate dose. Artemisia annua was especially useful when standard antimalarials are unavailable or unaffordable. One partner had been thanked by the local communities because this was the only source of antimalarial treatment available to them. Regarding pharmacovigilance, miscarriages were the only reported adverse event, occurring during the first trimester of pregnancy. Therefore Artemisia should be contraindicated in the first trimester of pregnancy.

An experimental ethno-archaeology on Qinghao in ancient Chinese medical texts (Dr Elisabeth Hsu)

“Qinghao” is first mentioned in a text found in a tomb dated to 168 BC where it was said to be used as a treatment for haemorrhoids or perianal ticks. It was said to “brighten the eyes”, and was used for wound healing. Malaria became more important in China with the spread of rice cultivation from about 1000 AD. From that time the imperial government recommended Qinghao more widely for malaria. Arsenic was also used to treat malaria. The traditional preparation involved soaking the plant in water and “wringing out” the juice, or pounding it with a pestle.

Ancient texts describe several species of Artemisia. Qinghao seems to equate to Artemisia apiacea, and Huang Hua Hao to Artemisia annua. However some botanists debate whether these two are the same or different species.

The Pharmacology of traditional Chinese preparations (Dr Colin Wright, University of Bradford)

Artemisinin was isolated from Artemisia annua in the 1970s. It is an unusual compound with a peroxide ring which is the key to its antimalarial properties. It is highly active against P. falciparum in vitro, and may reduce transmission by killing gametocytes. It is non-toxic. However recrudescence of parasites often occurs, there are concerns about resistance, and it is costly to produce. Therefore some NGOs have tried growing plants rich in artemisinin, harvesting and drying them, and using them as tea. This has been studied in clinical trials by Mueller (2000, 2004) and Blanke (2008). All these trials were on semi-immune African adults with Plasmodium falciparum, and in spite of initial efficacy, there were high rates of recrudescence. Furthermore the results cannot be extrapolated to young children who do not have such well-developed immunity as older children and adults.

We therefore tried to reproduce the methods described in the Chinese texts:

1. Soak in 0.4dL of water 2h + Wring out + Take juice in its entirety
2. Soak in 0.4dL of water 12h + Wring out + Take juice in its entirety
3. Pounding followed by squeezing out juice
4. Infusion of dried leaves 10g + 1L water

Samples were extracted with hexane and artemisinin content was measured by capillary gas-liquid chromatography. In vitro activity was tested against P. falciparum (strain K1) cultured in human red blood cells. Activity was assessed by measuring inhibition of incorporation of [3H]-hypoxanthine in the presence of extracts compared with that in untreated infected cells.

In vivo activity was studied using Peters’ 4-day suppressive test. Mice infected with Plasmodium berghei were given AA extract orally in one or more doses of 500L every 12 h. On the 5th day, a blood sample was taken and parasitaemia measured

Results:

1. Artemisinin concentration in different preparations

Method Artemisinin concentration (mg/L) Low yielding strain High yielding strain (extraction efficiency, %) Infusion (10g/L) Not tested 25.2 (54) Wrung Juice  (2 h soak) 24.6 45.9 (4.5) Wrung juice (12 h soak) Not tested 72.6 (7.3) Pounded juice Not tested 293 (15)

2. In vitro antiplasmodial activity (IC50): the juice appears to be 6-17x more potent than pure artemisinin, possibly due to a synergistic effect.

Method Dilution giving 50% growth inhibition Artemisinin concentration at 50% inhibitory dilution (ng/mL) 2 h wrung juice (Low artemisinin) 1:50,000 0.50 2 h wrung juice (High artemisinin) 1:200,000 0.21 12 h wrung juice (High artemisinin) 1:200,000 0.33 Pounded juice (High artemisinin) 1:500,000 0.59 Artemisinin - 3.80

3. In vivo: 2 doses of pounded juice had a similar effect to 1 dose artemisinin

Extract/Drug Dose Given every 12 h % Suppression of parasitaemia 2 h wrung juice (low/high artemisinin) 500 μL × 3 0 12 h wrung juice (high artemisinin) 500 μL × 3 (≡ 4.5 mg/Kg) 0 Pounded juice (high artemisinin) 500 μL × 1 (≡ 6 mg/Kg)   52 Pounded juice (high artemisinin) 500 μL × 2 (≡ 12 mg/Kg) 95 Pounded juice (high artemisinin) 500 μL × 3 (≡ to 18 mg/Kg) 96 Artemisinin 30 mg/kg 88

However these studies need to repeated. There was no follow up for recrudescence.

These results have been published:

Colin W. Wright, Peter A. Linley, Reto Brun, Sergio Wittlin and Elisabeth Hsu (2010). Ancient Chinese Methods Are Remarkably Effective for the Preparation of Artemisinin-Rich Extracts of Qing Hao with Potent Antimalarial Activity. Molecules 2010, 15, 804-812. http://www.mdpi.com/1420-3049/15/2/804/pdf

Systematic review of clinical evidence on Artemisia annua herbal preparations (Dr Merlin Willcox)

The systematic review of clinical trials on Artemisia annua herbal preparations has been updated to include two new clinical trials. The overall conclusion remains unchanged, that Artemisia annua preparations are safe and well tolerated in adults, but there is no published data on children or pregnant women. High-dose ethanolic extracts (tablets or capsules) seem to be the most effective but this needs to be replicated in Africa, in patients with falciparum malaria. Aqueous infusions clear parasites in about 70-80% of patients but there is recrudescence in 35-40% by day 28. Increasing the dose from 5g to 9g per litre has no effect.

Ethanolic extract of Artemisia annua: a promising new preparation (Dr Dirk Rezelman)

ACTs are too expensive for Africa, and are only made affordable by subsidies. Farmers are reluctant to invest in developing production due to uncertain market, so Artemisia production may fall.

Ethanol extraction is used in the current production of ACTs, but we propose to stop at the extraction rather than to go on to produce a pure compound. The dried leaves are a herbal substance whereas the ethanol extract is a herbal product. The extract has the advantages of increased absorption. There are also potential advantages of reduced auto-induction and synergy (which do not occur with the pure compound) but this has not yet been proven. This can then be combined with other drugs such as piperaquine to make a COMBined Artemesia Therapy (COMBAT). A 1:1 dilution of the extract in peanut oil was found to be stable at 20’C for at least 4 months. Stability at 40’C is not being investigated.

A pilot phase I crossover clinical trial was performed on 7 volunteers to determine the pharmacokinetic properties of the extract (containing 132mg artemisinin) compared to pure artemisinin (500mg). The Cmax was 239ng/ml and 315ng/ml respectively, and the AUC was 678ng and 1250ng respectively. Therefore the bioavailability relative to the dose received was 2.1x greater for the extract than for the pure compound. 

Plenary discussion: How can existing programmes be improved based on current knowledge? What are priorities for future research?

Keith (Anamed) pointed out that this depends on the aim of the programme – whether it is to have better drugs, or to improve local production. Can other preparations be used at local level in existing programmes?

Anthony Ellman responded that the aim is to look at each approach, what works, what is proven to work long-term. He cautioned that one needs to be careful about anecdotal evidence. What about recrudescence, risk of resistance? More clinical trials are needed, comparing herbal preparations with ACTs. ACTs have been proven to work and are funded by global subsidies, the Gates Foundation, and the A2S2 scheme funded by Triodos bank. Programmes on the ground should be keeping good records. Anthony wouldn’t tell Anamed it should stop its activities, and thought that the herbal preparation was not as great a risk as monotherapies for provoking resistance because of the combination of compounds it contains. He recommended that WHO should speak with Anamed.

Dirk suggested there would be a better chance of getting funding for a trial if combining the tea with another drug.

There is variability in artemisinin content even in commercial production all using standard production – content varies from 0.3 to 1.5%. Artemisia apiacea contained 0.05 – 0.21mg/ml, mean 0.12 mg/ml artemisinin. However Keith cautioned that there are 10 other antimalarial compounds in A. annua as well as artemisinin.

It is also important to consider how many people use the proper dose of pharmaceuticals, and what % of pharmaceuticals are fakes. Some reports from South-East Asia show that 38% of artemisinin tablets on the market are fake and contain 0% artemisinin! Shelly Burton pointed out that corruption is endemic and leads to distribution problems for modern drugs; a herbal preparation which can be grown and made locally will empower ordinary people.

Debate 1: Which preparations of Artemisia annua to promote in what circumstances? ACTs vs teas vs juice vs ethanolic extracts

Dirk’s opinion was that if ACTs are available, use them. If the stocks are limited keep them for children, and use herbal medicines for older children and adults. If there is nothing else, AA tea is better than nothing. Farmers should be encouraged to sell their Artemisia crops to an extraction unit, and to be given the standardised drugs afterwards. Henk agreed that Anamed should promote the cultivation of AA, but then sell the plant for making the extract locally.

Keith replied that the farmers would need to be paid a fair price, which currently they are not by the large companies extracting artemisinin. East Africa Botanicals pays $500 a ton for Aa dried leaves, which farmers feel is not enough. We need “fair trade” labelling for medicines! He also felt we must never discourage people from growing Artemisia in gardens. If there is a sudden stock-out of ACTs, people need to have a fallback. ACTs must be affordable and quality assured (as there are many fakes). It is better to foster self-reliance than dependence. When ACTs are available, people stop growing Aa – which implies people prefer to use ACTs. However when the ACTs run out they have nothing to fall back on. To treat everyone we would need more than 100 – 200m treatments a year, and this will increase the danger of resistance.

Elisabeth Hsu felt that the juice is still far removed from being introduced into practical projects, but it raises interesting research questions, such as whether resistance or recrudescence would occur with this preparation, and whether herbal preparations can empower local people to take more responsibility for their health care. Also research is needed on the different phenotypes and genotypes of the plant.

Jim Anderson reported that the Oxford Centre for Development Research takes a multisectoral approach, such as research on permaculture for helping AIDS patients. Social factors are part of research.

Dirk said that ethanolic extraction cannot be done on a small local scale, as it requires equipment and skilled people. A small factory could produce 40 million treatments a year. However each country could have its own production unit which should be controlled locally. The plants should be grown according to GAP and the medicines made according the GMP. We can achieve more aiming for a large crowd of patients. Henk felt we have enough knowledge now to go into action without more research.

After some debate Dirk and Keith arrived at the compromise recognising that there is a role for both of their approaches. Dirk’s standardised extract could be used to provide treatments for patients who can be reached by the health system infrastructure (provided it is proven safe in clinical trials), but the Anamed approach is very valuable in remote areas which have no access to any other form of healthcare. Anamed would in principle support the cultivation of AA both for local production of ethanolic extract and for local use as a herbal medicine, and would encourage further studies on the fresh juice. 

Debate 2: Combination therapies: For future trials should we combine Artemisia annua with conventional antimalarials or other herbal antimalarials and if so which)?

The rationale for using another herb or drug with AA would be to improve its efficacy, improve absorption, and so reduce recrudescence and the risk of resistance. Merlin started this debate by discussing combinations would make sense to test. These could include modern drugs (such as SP, AQ), or other antimalarial plants:

• Ascorbic acid increases solubility of artemisinin in water (Bilia et al (2002), J Pharm Sci)
• Turmeric: Curcumin shows synergistic action with Artemisinin in mice (Nandakumar et al, 2006)

Traditional Chinese Formulae (Elisabeth Hsu is planning to research what these were)

• Cinchona bark
• Cryptolepis sanguinolenta
• Argemone mexicana (grows in dry conditions )

Keith quoted the Anamed book which lists possible other antimalarial plants:

• Lemon grass
• Neem
• Papaya
• Vernonia amygdalina

Grapefruits are widely grown; their juice could be used as a source of Vitamin C and will inhibit autoinduction of enzymes which break down artemisinin. Dirk Rezelman suggested that piperaquine and naphthoquine are the best partner drugs as there is least resistance to them and they have not been used as a monotherapy in Africa.

Colin Wright mentioned that some plants and drugs can be used as resistance reversers, but the evidence for this is mainly in vitro. He suggested that different combinations should first be tried in vitro before going to clinical trials.

Debate 3: Will the widespread use of herbal Artemisia annua preparations and/or ACTs promote resistance to Artemisinin?

Merlin summarized the studies reporting the first cases of “resistance” to artemisinin in Cambodia (Noedl et al 2009, NEJM 361: 540; Dondorp et al 2009, NEJM). He presented the arguments for and against the hypothesis that use of herbal Artemisia annua may lead to the evolution of resistant parasites. He also presented arguments for and against the hypothesis that widespread use of pure artemisinin derivatives or ACTs may also lead to the evolution of resistance. Unfortunately there is no decisive evidence one way or the other, and no one has yet designed an experiment which would be able to prove or disprove these hypotheses.

Dirk Rezelman pointed out that neither of the studies used artemisinin. Dondorp et al used artesunate and Noedl et al used dihydroartemisinin (DHA). Dirk pointed out that in his pharmacokinetic studies he did not find any DHA in the blood of volunteers taking Artemisia extract, and therefore that contrary to popular belief, artemisinin is not metabolized to DHA. Furthermore there may be genetic differences between populations in the metabolism of artemisinin.

Elisabeth said that studies in Yunan which had attempted to induce resistance in vitro had found this was difficult to do. Artemisinin has been used for 30 years before possible signs of resistance have developed.

Anthony Ellman said that resistance is made more likely by monotherapy and subtherapeutic doses. A possible strategy to prevent resistance is to “carpet bowl” with ACTs to kill off all resistant parasites. Dirk said that if a whole plant extract was used at a higher dose there was a lower risk of resistance developing.

Colin Wright said that resistance is probably inevitable, and made an analogy with climate change. He suggested that instead of trying to prove whether or not resistance will develop, we should concentrate on finding ways to minimize the risk of resistance. For example a strain of Artemisia annua has been bred in York which contains 2% artemisinin, but it is not yet known whether this will work if scaled up.