The Dublin doctor who is not beating cancer


As I’ve said so many times before, this blog is my hobby. I write about what interests me for my own amusement. If it also interests you, that’s awesome. Fortunately, I’ve found that several thousand people a day do like what I lay down on a daily basis, sometimes with occasional spikes to ridiculous levels of traffic (such as this Dr. Oz post, which for some reason broke all traffic records for this blog since the very beginning, and not by a little bit) and, more recently, this post about how the CDC did not apologize and “admit” that the flu vaccine this year doesn’t work, which got quite a respectable spike, about 4-5 times normal daily traffic that took until today to drift back down to normal) but that’s not the primary reason I write. I did this blog back when its readership was minuscule in the months after its beginning, and I continue to do it now that it gets quite respectable traffic for a medical/science blog with a single blogger.

That being said, none of this means that I won’t on occasion take requests, particularly when more than one reader emails me about a topic. So it was that, over the last few days, I’ve had several of you send me a link to this book review, which appeared in the Irish press on Sunday, specifically It’s by the book editor (John Spain) and entitled The Dublin doctor who is beating cancer. It’s subtitled John Spain on what is probably the most important book to be published here this year. The book is entitled Stop Feeding your Cancer, and it’s by a Dublin GP named John Kelly. Spain characterizes it thusly:

It may seem an extraordinary statement to make, given the billions poured into cancer research by multinational pharma companies and medical research facilities. But a new book by a veteran GP on the northside of Dublin may well offer the first real proof that a way has been found to beat cancer.

This is, of course, utter nonsense. Depending on the cancer, there are already several ways to “beat cancer.” If you have early stage breast cancer, for instance, surgery combined with radiation ± chemotherapy ± drugs that block estrogen can “beat” cancer. Heck, even if it’s stage III cancer, patients have a reasonable chance of living 10 years. As long as the cancer hasn’t spread beyond the lymph nodes under the arm, breast cancer is potentially “beatable.” The same is true for other cancers. Lymphomas and leukemias can be cured with a combination of chemotherapy ± radiation (for lymphomas). Other solid tumors can be cured with a combination of surgery ± radiation therapy ± chemotherapy. This book review has thus started out with a breathtakingly ignorant statement. That does not bode well for the quality of the rest of the review or how well Spain understands cancer. Such is life.

Spain continues:

In Stop Feeding your Cancer, Dr John Kelly does not produce a “cure” for cancer. Instead he presents convincing evidence, based on the experience of his own patients, which shows that cancer can be stopped in its tracks and even reversed into a dormant state, allowing sufferers to regain good health and lead normal lives.

The core of this approach is the discovery a decade ago of a direct link between the consumption of animal protein and the development of cancer. Cancer cells need protein to divide and flourish. Cut off the supply of animal protein and you can stop the growth and spread of cancer cells. You can starve the cancer into submission.

So right off the bat we know that Kelly has not done anything resembling a clinical trial. As so many promoters of medical pseudoscience have done, he’s collected anecdotes from his practice. As they say, when it comes to medicine, the plural of “anecdote” is usually not “data,” and this sounds like an excellent example of this phenomenon. But first, why does Kelly think that cutting out animal protein will treat cancer? According to Spain, Kelly based this idea on the China Study.

I don’t believe I’ve ever written about the China Study before, but Harriet Hall has, not just once but twice. It’s an epidemiological study of diet and health conducted in villages throughout China by Colin Campbell of Cornell University. It was first published in the US in 2005 and has sold over a million copies, apparently. It’s also a favorite of advocates of “raw food” and vegan diets as a panacea because the major claim made in the book is that we can prevent or cure nearly all diseases (heart disease, cancer, diabetes, autoimmune diseases, bone, kidney, eye and other diseases) if only we would give up meat and dairy products entirely, drastically decrease our protein intake, and eat a strictly plant-based diet. You can see why such a view would be attractive.

Unfortunately, Campbell’s book, aside from being a book rather than a study subjected to the peer-reviewed scientific literature, has a lot of problems, as related by Harriet Hall: References whose conclusions are not as represented in the book and do not support Campbell’s thesis, including a citation that touts the quackery that is the Gerson protocol; downplaying and even failing to mention research whose results do not agree with the China study; and finally shoddy research. Indeed, a reanalysis of the raw data of the China study by Denise Minger found many weaknesses and errors, including finding no direct correlation between animal protein intake and cancer and many others. Meanwhile, a lot of follow up research failed to validate the findings reported in the China study. Seriously, Spain (and Kelly) really need to read this roundup of science-based analyses and criticisms of the China study. Spain, for instance, is completely unskeptical of the claims made in the China study and swallows what Kelly says about it, who in turn swallowed what Campbell claimed.

So, basically, right off the bat, Kelly is proceeding from a flawed premise. It’s just not as simple as Campbell or Kelly makes it out to be and there exists, either in the China study or elsewhere, strong evidence that a vegan diet can cure cancer. Unfortunately, Kelly’s survey of his patients doesn’t provide such evidence either, although you wouldn’t know that from Spain’s description of the book:

And the results, which he details in his book Stop Feeding your Cancer, are jaw dropping. As in any large GP practise in Ireland, a number of Kelly’s patients develop cancer every year. He began to tell them about the link, gave them a copy of The China Study, and suggested they go on an animal protein free diet (no meat, no dairy produce). At the same time he continued to refer them on to cancer specialists in the normal way and did not try to dissuade them from having whatever surgery, radiotherapy or chemotherapy might be recommended.

What emerged is truly extraordinary. All of the patients who adopted the animal protein-free diet and stuck to it strictly found that their cancer stopped growing and spreading. The tumours became dormant, sometimes even reducing in size.

Since all of these patients were also having conventional cancer treatment, it was not possible to categorically say they had been saved by the diet alone. But conventional cancer treatment alone does not have this success rate and in many cases the specialists involved were amazed at the recovery of the patient. (Some of Kelly’s patients had told their specialists they were on the diet, others did not. Invariably, the specialists put the recovery down to surgery or chemotherapy.)

Kelly gives the full details of over half a dozen cases in the book, including profiles of the patients and their lifestyles. The cases cover prostate, lung, colon, bowel, brain and other cancers. In some cases the patients stuck rigidly to the diet; in others they became complacent as they got better and could not resist going back to having steaks and fry-ups when their specialists gave them the all clear – whereupon their cancers came back.

Let’s see. We have no inclusion criteria other than cancer and apparently no exclusion criteria. The patients were treated normally. Are there any statistics on these patients? What stages? What cancers? What treatments? Are these all detailed systematically? Was there approval from an ethics board, the Irish equivalent of institutional review boards (IRBs) in the US? After all, this “field trial run,” as Kelly describes it, is basically a clinical trial. How do we know that this isn’t simply a massive case of confirmation bias, very much like when the HomeFirst Clinic in Chicago, a practice that does not vaccinate, famously claimed that it had no autistic children in its practice. Mighty convenient—isn’t it?—that the only patients who seemingly suffered progression of their cancer were ones who had “fallen off the wagon” as far as their vegan diet goes. Particularly telling is this passage:

The personal stories he gives are very human as patients try to stick with the diet, the disease comes and goes and the patients swing between despair and elation. Invasive treatments run alongside attempts to keep to the diet.

That sure sounds as though patients didn’t uniformly do well on the diet and that their difficulties with the diet are being used as a blame-the-victim strategy. I don’t think it’s being done intentionally, but rather as a means of rationalizing why not every patient on the vegan diet did as well as Kelly apparently expected. All of this makes Spain’s description of him as “scrupulously scientific” as doubtful at best, although the description of Kelly as “open-minded” brings to mind the crack about being so open-minded that your brains fall out.

Patients who didn’t stick to the diet weren’t the only patients who didn’t do well. Not surprisingly, given its deadliness, patients with pancreatic cancer weren’t miraculously cured by switching to a vegan diet. (Go figure.) Yet Kelly has an explanation for why patients with pancreatic cancer didn’t do well. Unfortunately, it is an explanation that is neither convincing nor rooted in reality. When I read this review, I had a hard time believing that Kelly doesn’t realize how physiologically nonsensical his explanation for why vegan diets won’t cure pancreatic cancer is:

There is one exception to this, pancreatic cancer, the virulent form that killed Brian Lenihan and is usually fatal. But far from undermining the diet theory, Kelly, who has lost several patients to pancreatic cancer including his own brother, says this is actually an exception that proves the rule. The reason the animal protein free diet does not work against pancreatic cancer and some gastric cancers, he explains, is that the pancreas itself produces animal protein. The metabolising enzyme involved called Trypsin may hold the key to solving this problem and Kelly calls for research into anti-Trypsin drugs.

Huh? Trypsin is no more or less an “animal protein” than any other protein in the human body, given that human beings are animals and their proteins are very similar, in some cases close to identical, to the “animal proteins” in food. By Kelly’s criterion, pretty much any enzyme in the body is an “animal protein.” An interesting twist to this truly silly explanation is that Kelly thinks trypsin generates these “animal proteins.” All trypsin does is to cleave proteins at specific amino acid residues. Specifically, trypsin cleaves proteins at amino acids lysine or arginine, except when either is followed by proline. That’s all it does. It doesn’t matter if the proteins are plant or animal proteins. Trypsin cleaves big proteins into little peptides that can be further broken down in the small intestine.

Even stranger, Kelly seems to think that antitrypsin drugs would somehow reverse this production of “animal protein” by trypsin. This is curiously the exact opposite of the quack Gonzalez protocol, which blames pancreatic cancer on a deficiency of pancreatic enzymes and claims to treat pancreatic cancer (in part) by replenishing those enzymes in many supplements. It’s basically the opposite kind of cancer quackery as the Gonzalez protocol, but based on the same sorts of ideas: that pancreatic cancer is caused in part by dysfunction of pancreatic enzymes. It’s also profoundly dumb to anyone with a basic knowledge of GI physiology. Kelly should be ashamed of himself as a physician for having so poor a grasp of basic pancreatic biochemistry and physiology.

So why did Kelly write such a wrong-headed book? Well, here:

In the book he calls for such research; in fact his main motivation for writing the book is the disinterest he has faced from consultants and oncology specialists in the past few years who are dismissive of the diet theory to the point of being insulting. They stick with their surgery and radiotherapy and chemotherapy even though in many cases it either does not work or offers only limited time. They regard a dietary solution as simplistic or as flaky alternative medicine. What would a mere GP know? After all, they are the experts.

Ah, yes, the Brave Maverick Doctor, shunned by physicians who actually understand the science of cancer because of his brilliance. Did it ever occur to Spain (or Kelly) that perhaps the reason that oncologists are dismissive of Kelly’s ideas that animal protein is the cause of cancer and a vegan diet is the cure is because they are divorced from science. It’s actually not unreasonable to hypothesize that a vegan might prevent cancer. It’s a testable claim, although the existing evidence base is somewhat conflicting and the very tome upon which Kelly bases his hypothesis is shaky at best, biased and cherry picked at worst.

Unfortunately, Kelly did nothing substantive to test his claim. Single anecdotes are notoriously subject to all the fallacies of thinking to which the human brain is prone. Bundling a bunch of anecdotes together without pre-specifying inclusion and exclusion criteria and outcome measures (at least) does not constitute a case series. Spain wouldn’t be expected to know that, but Kelly sure as hell should. If he was really interested in testing whether a vegan diet could cure cancer (or at least greatly slow its progression), he should have published his results in the peer-reviewed medical literature, not as a book.