Holford Watch: Patrick Holford, nutritionism and bad science

There have been lots of exciting news stories about turmeric over the last few years: from claims that it might slow Alzheimer’s Disease, to reducing the incidence of leukaemia, or being a potent weapon against several cancers and cystic fibrosis. There are the usual claims that it has general anti-inflammatory effects and modulates the immune system although the mechanisms are presently unclear. There is a good overview about curcumin and some of its principal researchers in a recent Scientific American: Spice Healer.

In his BMJ Rapid Response Holford eulogised the benefits of curcumin (the active ingredient of turmeric that has attracted most interest) and made a plea for a proper sense of perspective.

The most recent review on turmeric, in the January issue of the Journal of Clinical Immunology (5), states: “Traditionally known for its an anti-inflammatory effects, curcumin has been shown in the last two decades to be a potent immunomodulatory agent that can modulate the activation of T cells, B cells, macrophages, neutrophils, natural killer cells, and dendritic cells. Curcumin can also downregulate the expression of various proinflammatory cytokines including TNF, IL-1, IL-2, IL-6, IL-8, IL-12, and chemokines, most likely through inactivation of the transcription factor NF-kappaB. Interestingly, however, curcumin at low doses can also enhance antibody responses. This suggests that curcumin’s reported beneficial effects in arthritis, allergy, asthma, atherosclerosis, heart disease, Alzheimer’s disease, diabetes, and cancer might be due in part to its ability to modulate the immune system. Together, these findings warrant further consideration of curcumin as a therapy for immune disorders.”

The safety data on turmeric is exceedingly good. Even if somewhat premature, is it really such a sin to recommend people in Britain to eat this spice on a regular basis, as done in Asian countries where prostate cancer incidence is exceedingly low?

Should readers feel abashed? Not really. Holford is not merely recommending that people should eat turmeric on a regular basis. He is aware that it isn’t practical for most people to obtain the recommended curcumin intakes by eating “this spice on a regular basis”. Presumably, this encouraged him to develop and promote his own curcumin supplement and advise an intake of up to 4g per day which is around an estimated 20 times the amount consumed in the average indian diet.

Registered Dietitian, Catherine Collins, disputed the practicality of Holford’s recommendations in her own response to him.

Registered Dietitians would disagree with nutritional therapist Patrick Holfords example of curcumin as a potent agent against disease, despite the immunology primer provided as evidence.

Epidemiological and limited clinical evidence support a potential role for phytochemicals as human cancer chemopreventive agents (1). The effects of polyphenols - such as curcumin - are more pronounced in vitro, using high concentrations which are not physiological in vivo (2). Clinical studies confirm the poor bioavailability of curcumin (3)…

A ‘clinical’ dose of 3.6g curcumin is sufficient to achieve in vivo biochemical influence (5). Turmeric powder – a constituent of curry powder - contains 3.1% curcumin by weight (6), so ‘research-to-recipe’ equivalence would require consumption of 110g of turmeric powder daily. Mr Holfords exhortation ‘let the public eat curry’ is thus nutritionally irrelevant in the context of current research… [Emphasis added.]

If the public can not reasonably eat enough curry to reap the purported benefits of curcumin, then it might seem logical to advise them to take a supplement. However, in two interesting posts, researcher Abel Pharmboy discusses some relevant issues:

1. Curcurmin for Cancer Part 1 asks the question, “Do concentrations of curcumin (or any compound) with anticancer effects in cell culture actually occur in the bloodstream of patients who take dietary supplements containing the compound?”.
2. Curcurmin for Cancer Part 2 discusses the limited literature on the actual bioavailability of curcumin and explores whether “the [recommended] doses of an herb or dietary supplement may not achieve sufficient concentrations in the body to mimic the seemingly miraculous effects observed in cell culture studies”.

Abel Pharmboy’s scrutiny of the literature leads him to conclude that because of the way that curcumin is absorbed in the body, it would presently be necessary to take a massive dose, every hour (presumably until such time as somebody develops a time-release formulation) along with another agent to boost bioavailability. Even then, increasing a dose does not always increase the concentration in the blood levels. Theoretical calculations based on how much curcumin makes it into the bloodstream from an oral dose plus the theoretical dose necessary for an anti-cancer effect, indicate that you would need an hourly dose of between 40-200 grammes.

[T]here is no way to know how much of this curcumin product must be taken to achieve blood concentrations consistent with cancer cell growth arrest or cell killing.

So, yes, curcumin might be a promising anticancer compound but only if you could literally shovel quantities of it into the bloodstream. So, what should be done? Remember, I do this kind of work for a living and am a big proponent of natural products as a source of anticancer drugs… [Emphasis added.]

Abel Pharmboy concludes:

In the meantime, to continue marketing curcumin, even in a highly-bioavailable form, is misleading and gives false hope to patients.

At the very least, it is a tremendous waste of money.

In his Scientific American piece, Gary Stix alludes to the fact that curcumin does not have a wholly positive news flow. There are some questions about interactions with herbs, dietary supplements and drugs. Any substance that has as many possible biological impacts as those ascribed to curcumin may also be capable of unwanted biological actions. Stix points to research that indicates “the possibility that this spice may sometimes actually encourage the survival of cancer cells”.

Much of the present data about curcumin has been obtained from cell studies in laboratory settings or animal studies. Several interesting human trials are in progress that will provide more information and give a fuller picture of the action of curcumin on human physiology or pathology.

Marketing materials that emphasise relevant scientific terms can be very persuasive. The material for Holford’s curcumin supplement contains terms that will be familiar to people with a variety of illnesses, including cancer:

Research supports the wide-ranging beneficial properties of the curcuminoids, and especially curcumin. These include powerful antioxidant properties, modulation of the production of inflammatory signal molecules (prostanoids and leukotrienes), and inhibition of the growth of cells which have not properly differentiated, or whose growth pattern is abnormal. Because curcuminoids act through both intervention and preventative means, they are accordingly considered among the most potent of the known antioxidants.

It should be remembered that in his BMJ Rapid Response, Holford criticised Ben Goldacre because:

he attacks advice to eat turmeric for cancer protection, including that of the prostate…

However, although Holford is willing to disseminate the research news about turmeric with some enthusiasm, he has not commented on the research that questions:

1. whether curcumin might stimulate the growth of some cancers
2. the bioavailability of curcumin in supplements
3. the implausibility of being able to take/afford sufficient curcumin to be clinically effective on a range of disorders

According to Catherine Collins (as quoted above):

The interpretation of clinical nutrition research and its application to prevent, treat or manage disease across individuals and populations should be confined to those capable of the practice.

Currently available supplements do not provide information about their bioavailability and its relationship to therapeutic doses for a number of implied symptoms and conditions (there are some weak advertising regulations that restrict the claims that can be made). If current and future research confirms the potential clinical significance of curcumin, the people who market it will probably need to research derivatives that would have the necessary bioavailability and can be proved to target relevant biological mechanisms that are involved in particular symptoms or illnesses. For now, it seems that there is no scientific basis to support Holford’s advice to take curcumin supplements.

Update July 8 2008: We are working on an update of our overview. Abel Pharmboy is currently working on an update to his posts about curcumin. In the interim he says:

My biggest concern is not so much that curcumin won’t work but rather that it is being formulated with piperine (under the brand name Bioperine here in the States) to increase curcumin bioavailability. This effect could also prolong the action of other drugs the person may be taking, increasing their chances of adverse effects.

…I haven’t yet seen any new data on the bioavailability issue but I am deeply concerned about attempts to modulate endogenous drug metabolism pathways to enhance the action of curcumin.