I’ve already devoted more time to Protandim than it deserves. I’ve written about it twice on SBM: here and here . But I can’t resist covering a new Protandim study that not only serves as a bad example but that made me laugh.
Protandim is a mixture of 5 herbal supplements intended to upregulate the body’s own production of antioxidants. Its patent application claimed that it was useful to treat or prevent an astounding 126 diseases and medical conditions, from tinnitus to aging, from hemorrhoids to cancer. At the time of my last article, only one human study had been done. It found increases in blood test markers and interpreted them as a surrogate for increased antioxidant activity in the body, but did not even attempt to assess whether those increases corresponded to any measurable clinical benefit, for cancer or for anything else. I begged Protandim supporters not to ask me about it again until there were human clinical studies with meaningful outcomes.
Now there is finally a second human study, although still not one that qualifies as a clinical trial. Curiously, it is not listed on the company’s website. I wonder why? Perhaps because it showed Protandim didn’t work. Oops.
Most customers seem to be taking Protandim because of its alleged ability to prolong life, improve diabetes, prevent cancer, make them feel better, etc. Instead of studying any of those things, the researchers chose to study alveolar epithelial permeability.
In the introductory section of their paper, they explain that alcohol use disorders (AUDs) have been linked to the development of acute lung injury. They admit that the underlying factors are elusive and probably multifactorial. One potential contributor is the patients’ propensity for severe pulmonary infections mediated through abnormal pulmonary innate immunity. Additionally, they say that prior research had “suggested” an association between AUDs and abnormal alveolar epithelial permeability which “might” be mediated through pulmonary oxidative stress. So they did a double blind study, not to test whether Protandim would improve lung disease or reduce the rate of infections in alcoholics, but to see if Protandim would reduce markers of oxidative stress and normalize abnormal alveolar epithelial permeability.
To recap their chain of reasoning: alcoholics might develop lung disease, that lung disease might be correlated with abnormal epithelial permeability, protein levels measured by bronchoalveolar lavage (BAL) might be a valid measure of permeability, permeability might be affected by underlying oxidative stress, and Protandim might reduce oxidative stress by stimulating the body to produce its own antioxidants. Do they perhaps think that lots of “mights” add up to a “mighty” argument?
Why would they want to study this particular mixture of 5 herbs? The second listed author, Joe McCord, has a vested interest: he is an officer of the LifeVantage company, the manufacturer of Protandim. They explain that Protandim is “a nutraceutical with a lengthy history of use in homeopathic, Ayurvedic, and traditional Chinese medicine.” An interesting statement, since Protandim was invented only a few years ago by a person with no medical background and it was patented in 2007. Doubly interesting since it belies the common myth that natural medicines are not profitable because they can’t be patented.
How many Protandim customers are alcoholics taking it for lung injury due to alcohol abuse? I would guess not many. Why on earth would they pick an esoteric detail like this to study and why would they look at Protandim’s influence on lab tests instead of looking for a useful clinical benefit?
Bronchoalveolar Lavage (BAL)
They randomized 30 alcoholics and gave 16 of them placebo and 14 of them a double dose of Protandim. There are some methodological problems: randomization was imperfect, with unequal numbers and some significant differences between the groups; some possible confounders were not eliminated (for instance, some of the subjects got intermittent benzodiazepines for alcohol withdrawal symptoms); and some of the blood tests were only done on smaller subsamples of the group. A separate small control group of 11 normal subjects also got a single BAL and their BAL protein levels were not significantly different from the AUD subjects, which sheds doubt on the whole rationale for the study by failing to confirm that alveolar permeability correlates with AUD. They did BAL twice on the AUD subjects, before and after a week’s treatment with Protandim or placebo. The utility of BAL is limited, since there is a large range of normal values for the substances measured and a number of confounding factors. Of more concern, this procedure is invasive, unpleasant, and far from benign. It requires the use of drugs for topical anesthesia and conscious sedation and can result in serious complications. It involves passing a bronchoscope deep into the lungs, wedging it into a lung segment, squirting in a small amount of saline, and then sucking out the liquid for analysis. In other words, putting water into the airways: sort of a partial drowning. With known risks and no possible benefit to patients, why did the Institutional Review Board approve it? I hope their consent forms were adequate, and I hope they paid their subjects well.
They assessed alveolar epithelial permeability by measuring the total protein in bronchoalveolar washings. Total protein levels did not change in either experimental group. They also found no change in oxidative stress indices, epithelial growth factor, fibroblast growth factor, interleukin-1β, interleukin-10, liver function tests, or other blood chemistry tests. The one finding that was statistically significant was a significant decrease in plasma thiobarbituric acid reactive substances (TBARS), a marker of lipid peroxidation — but that was only in the placebo group!
In short, Protandim was significantly (p<0.01) worse than placebo. No wonder they’re not bragging about this study!
I really don’t get it. Why did they do this study? Why did they use twice the recommended dose? What was the IRB thinking? Why didn’t they study something with a useful clinical endpoint? As an e-mail correspondent said, “They make claims about diabetics being able to go off of insulin or reduce insulin…why not do a trial on that?” That’s an excellent point: a diabetes trial would not involve invasive procedures and would be far easier to carry out and far more meaningful. When advocates do esoteric, convoluted laboratory studies instead of straightforward simple clinical trials, it raises the suspicion that they believe at some level that such clinical trials wouldn’t help their case.
No, on second thought, I think I do get it: they want to prove, by any means possible, no matter how circuitous or far-fetched, that Protandim does something, anything, antioxidantish (not a word? Well it is now!). Then they can dredge the literature for every condition where a possible correlation with oxidative stress has been mentioned. Then they can try to convince customers they should take Protandim for all those conditions.
Big Pharma gets a lot of criticism, but aren’t Big Supplement and Big Multi-Level Marketing every bit as guilty of self-interest, distortions, and profit motives? At least Big Pharma can’t make its big bucks without first demonstrating effectiveness and safety to the FDA with clinical trials.
Does Protandim provide any real benefit to its customers? I don’t know, and they can’t hope to know unless they do proper clinical studies.
This article was originally published in the Science-Based Medicine Blog.