Never Mind the Bullocks, Here's the Science (18 page)

Opiate Variability

Opiates are a good example of a class of drug whose levels have to be juggled. Human beings have used opiates to relieve pain for many thousands of years. Today, besides the original drug opium, the opiate family includes morphine, codeine, oxycodone and heroin.

At any given moment, there are ten million people taking opiates for pain relief from cancer—indeed, 80% of people with advanced cancer experience great pain. But opiates are also administered for non-cancer-related pain, e.g. chronic pain and postoperative acute pain.

Opiates are generally good at relieving pain—but they are not perfect.

First, there is a relatively narrow margin between the dose that relieves pain and the dose that kills you via respiratory depression.

Second, there is enormous human variation in the dose needed to relieve pain. In one study of 3,000 people having hip-replacement surgery, the dose of opiate needed to provide adequate pain relief varied by a factor of 40 to 1. This variability is astonishingly large.

Many factors control this variation in the amount of pain relief in response to a given dose of opiate. These factors are also responsible for the variability in undesirable side effects.

In 15% of people the target site for the opiates varies, thus providing poor pain relief.

My Genes and Morphine

A few years ago, I had an accident with my 9 ft 6 in (2.9 m) surfboard. A wave smashed it into my left shoulder. (Never let your board get between you and a wave.) The ‘ball’ on my upper arm bone that fits into the ‘socket’ of my shoulder was smashed into about 40 pieces of rubble. Amazingly, the ball stayed intact, held together by clotting blood. The pain was so great that I didn’t really feel any pain – I was totally dissociated from my body.

It’s a Bad Sign in a hospital when staff start calling each other over to have a look at your X-ray. It’s even worse when they start whistling in astonishment, and then poke a curious head through the curtains to see just what you look like. After more than an hour in the Emergency Room cubicle, and half-a-dozen staff checking me out, the pain in my shoulder began to build up.

The doctor offered me morphine for the pain, and I said yes. He warned me that it might cause nausea and vomiting, and gave me a drug to suppress these unpleasant side effects of nausea and vomiting. Then he gave me the morphine.

I spent the next five minutes feeling very nauseated and wanting to vomit. I then became unconscious.

That was my Great Morphine Experience. I guess that I don’t have the genes to be an addict and write offbeat novels like the opiate-addicted author William S. Burroughs.

In other people, the protein that transports opiates operates so weakly that the morphine remains in the brain, causing massive respiratory depression, as well as nausea and vomiting. In another group of people the enzymes vary. For example, codeine works only because an enzyme in the body converts it to morphine. About 10% of people do not have this enzyme and, therefore, get no pain relief from codeine.

On-Demand Administration, where the patient self-administers the opiate dose via an intravenous delivery system until the pain is managed, has proven to be a big success. This is not surprising when you consider how huge the dosage range of these painkillers can be.

Other Drug Variability

In some cases, genetic variability means that drugs can kill a small percentage of people. This can happen with very powerful drugs that are used to treat very serious conditions.

Consider the drug 6-mercaptopurine, sometimes used to treat child leukaemia, severe rheumatoid arthritis and graft rejection. It is normally broken down by an enzyme in the body. But 1 in 300 people do not have this enzyme. In this case, the normally life-saving drug can kill.

At the moment, we do not have any simple way to find the one person in 300 who does not have the enzyme to break down 6-mercaptopurine. There are two tests for this. Both are quite different—one is a biochemical test, while the other is a DNA test. This is all still rather new. Even if patients take the DNA test, we are still not sure how to predict confidently what adjustments in dose need to be made.

Consider the drug abacavir, used to treat HIV/AIDS. In some
4-6% of people it causes a hypersensitivity reaction (HSR), and can sometimes kill. This HSR follows a well-described pathway—taking, on average, 11 days to become apparent, with 90% of patients presenting with symptoms in the first six weeks.

However, the test to identify the 4-6% of people who are hypersensitive to abacavir is much more satisfactory than the test for 6-mercaptopurine.

Consider the drug warfarin—an anti-clotting drug, often given to people to prevent the further formation of clots after a heart attack or a stroke. It is notoriously difficult to administer. Many medications, and even some foods, interfere with its activity. It also has a narrow therapeutic range. Too little will not prevent clots, but too much can cause bleeding and haemorrhage. The dosage needed to get the same results can vary by a factor of ten from patient to patient.

For several years we knew that two genes—Cytochrome P450 2C9 and the Vitamin K Epoxide Reductase Complex Subunit 1—were involved in its activity. Finally, a study in 2008 found that if these genes were taken into account in working out the dose of warfarin, the patients fared much better. The patients who benefited the most were those at the extremes—those who took the smaller, or larger, doses of warfarin. These patients—those who took less than 21 mg or more than 49 mg of warfarin per week—made up 46% of the 4,300 patients in the study. However, the jury is still out on the usefulness of these DNA tests, because there is still no easy algorithm to allow the doctor to use the result of the tests to predict what change in warfarin dose is needed.

But then in February 2009, another gene involved in the metabolism of warfarin was discovered. Presumably, once this newly discovered gene is taken into account, the treatment of patients with warfarin will improve again.

Direct-to-Consumer DNA Tests

If you wish to spend the money ($1,000 or so), you can send a sample of your saliva to a company that will, in return, send you back a ‘genomic profile’. While the test report may state that you have an increased risk of heart disease or diabetes, it may also tell you that ‘the test is not a clinical service to be used as the basis for making clinical decisions’. Welcome to the new world of DTC DNA, or Direct-to-Consumer DNA tests.

There are a few things to realise about these ‘genomic profiles’.

First, they do not look at all of your DNA, but at much less than 1% of it. Which bits are looked at will change the consumer’s risk profile (please note, the person who is paying for this so-called medical service is known as a ‘consumer’, not a ‘patient’). A good example of this is Type 2 Diabetes. Some 25% of Europeans will suffer from this disease during their lifetime. In 2007, there were fewer than ten DNA markers known for Type 2 Diabetes. Based on those markers, the consumers would have been sold a risk assessment. But since 2007 there have been two major updates in our knowledge of Type 2 Diabetes, with each update adding more DNA markers. After the first update, ‘more than 11% of people went from being told that they were at above-average risk to below average or vice versa; after the second update more than 10% were similarly reclassified’. In other words, we still know too little about DNA to understand what these companies are selling.

Second, ‘many of the diseases listed by the direct- to-consumer testing companies (e.g. diabetes, various cancers and heart disease) are so-called complex diseases thought to be caused by multiple gene variants, interactions among these variants,
interactions between variants and environmental factors’. In other words, we still do not fully understand how these diseases work. But the DTC testing companies will ignore the thousands of possible interactions. They will sell you the information that you
probably
have one or two genes that could increase your risk of a specific illness.

Third, no test made by human beings is perfect, hence the use of the word ‘probably’ in the sentence above. For example, there is a genetic test for Bipolar Disease. But it’s not a very good test. So, unfortunately, this test will wrongly report that more than 80% of those who actually do have Bipolar Disease do not have the supposed single gene for this disorder. And, on the other side, many people will be reported as having the gene when, in fact, they do not.

Fourth, we really still do not have the Big Picture of DNA. Human beings have about 20,000-23,000 genes. This is about the same number of genes as a pinot noir grape! So, obviously, we still have a very primitive understanding of exactly what a ‘gene’ is, and we clearly have a long way to go.

Fifth, these are still early days for DTC DNA testing. We have found only a minuscule fraction of the genes involved in various diseases. The vast majority of them are still to be found. Suppose that you pay the $1,000 for a genomic test, and the report states that you do not carry any genes associated with the likelihood of having a particular disease when you actually do carry the genes. Then you might decide that you can indulge in risky behaviour, your life becoming worse for having taken the test.

In early 2008,
The Observer
newspaper in the UK interviewed psychiatrist Professor Nick Craddock of
Cardiff University. He said, ‘These tests will only worry, confuse and mislead the public and patients. There is a long way to go before we have genetic tests that may be helpful to patients. Using tests at the moment is only likely to cause harm.’ In other words, our understanding of DNA is still very poor. There is absolutely no provision by the companies making money from DTC DNA testing to provide appropriate support and counselling for their patients (or ‘consumers’). Professor David Collier, of the Institute of Psychiatry, London, said much the same thing: ‘At best, these tests are clinically useless. At worst, their results could cause serious worries for patients.’

Future Horizons

Soon we will be able to map a person’s entire DNA cheaply and, more importantly, be able to understand what we find. We will be able to use this knowledge to work out who can benefit from a particular drug. Ideally, one day we will be able to identify the ‘responders’ (those who benefit from a particular drug) with a cheap and quick genetic test—and, at the same time, identify those who would have undesirable side effects.

Think back to the 90% of drugs that are sidelined because they are toxic to a tiny percentage of the population. In the future, we hope to be able to use these drugs only on the 99.99% of people who can benefit from them, and steer these drugs away from the 0.01% of people to whom they are toxic.

But before this can happen, we have a lot to learn about genetics.

Sir William Ostler said it all way back in 1892: ‘If it were not for the great variability among individuals, medicine might as well be a science and not an art.’

References

Aldhous, Peter, ‘Genes that tell an ever-changing story’,
New Scientist
, 1 August 2009, p 12.

Burke, Wylie, et al., ‘Translational genomics: seeking a shared vision of benefit’,
The American Journal of Bioethics
, March 2008, Vol 8, Issue 3, pp 54-56.

‘Discriminating on genes’,
Nature
, 5 July 2007, p 2.

Hunter, David J., et al., ‘Letting the genome out of the bottle—will we get our wish?’,
The New England Journal of Medicine
, 10 January 2008, Vol 358, No 2, pp 105-107.

International Warfarin Pharmacogenetics Consortium, ‘Estimation of the warfarin dose with clinical and pharmacogenetic data’,
The New England Journal of Medicine
, 19 February 2009, Vol 360, No 8, pp 753-764.

McKie, Robin, ‘Internet gene tests provoke alarm’,
The Observer
, 3 February 2008.

Reynolds, Kristen K., et al., ‘Individualizing warfarin therapy’,
Personalised Medicine
, February 2007, Vol 4, No 1, pp 11-31.

Roses, Allen D., ‘Genome-based pharmacogenetics and the pharmaceutical industry’,
Nature Reviews, Drug Discovery
, July 2002, pp 541-549.

Somogyi, Andrew A., et al., ‘Pharmacogenetics of opioids’,
Clinical Pharmacology & Therapeutics
, March 2007, Vol 81, No 3, pp 429-443.

Meat Rots in Gut
(An Indigestible Tale?)

Intellectually, I love vegetarianism. In fact, I was a vegetarian for many years. On the Big Scale, it’s absolutely true that if you want to feed an entire population, vegetarian diets consume fewer resources than meat diets. And further, on the Small Scale, it’s well known that a properly devised vegetarian diet can deliver real health benefits, including lower rates of obesity, heart disease and some cancers.

But I don’t agree with claims that human beings were never meant to eat meat. Nor do I agree with those who reckon that meat will putrefy and rot in your gut, giving off toxins that cause many diseases, including insanity, premature ageing and (God Help Us) enfeeblement.

Vegetarianism Good for the Planet?

In 1997, at the 24-26 July meeting of the Canadian Society of Animal Science in Montreal, David Pimentel, Professor of Ecology in Cornell University’s College of Agriculture and Life Sciences, said: ‘If all the grain currently fed to livestock in the United States were consumed directly by people, the number of people who could be fed would be nearly 800 million.’

However, sometimes livestock can graze on land that is too marginal to support agriculture. But most of the time this is not the case.