by senior futurist Richard Worzel, C.F.A.
Suppose you were a high school teacher who had just taken a job at a new school, and that you agreed to take over as the coach of the basketball team. You look at your new school’s team record, and find that it’s pretty mediocre, so decide you want to try something new. New school, new ideas – why not?
After looking around and thinking things over, it occurs to you that perhaps there’s a relationship between the kinds of shoes the players wear and how well a player does – and hence how the team fares. After all, a casual observation of the kids on the floor seems to show that those who have better shoes score more.
Since you have such a small sample of kids, you reach out online and get data from dozens of schools around the country, outlining what kinds of shoes their players wear, and how well they perform.
Once you start to analyze the data, the results are confusing. Some schools provide shoes, which are all the same brand, and the players do particularly well, but other schools that do the same do not. Some brands seem to have a slight correlation with good performance, but nothing that’s really enough to hang your hat on. There are hints of some kind of relationship, but nothing really powerful. In confusion, you take your results to one of the science teachers, whom you have befriended since your arrival.
She looks over the data, asks a couple of questions, then says, “Why don’t you look at how tall the players are instead of what shoes they wear?”
In my view, medical research is like that.
Is Butter Bad for You?
You’ve undoubtedly seen conflicting reports about food, for instance. Butter’s bad for you, says one, because it has saturated fats that increase bad cholesterol; you should stay away from butter and eat margarine. No, says another, margarine contains trans-fats and chemicals; butter’s natural and better for your body, so you should eat butter. You’re both wrong, says a third study: cholesterol is irrelevant, it’s all been blown out of proportion, so it doesn’t matter whether you eat butter or not.
None of these kinds of studies talk about the relationship between individual genetics and the studies that are performed. Yet, in my opinion, that’s like ignoring the height of basketball players in trying to identify the factors that relate to success in the game.
Suppose, for instance, that one group involved in a study on the effect of butter accidentally happens include a lot of people who respond badly to butter, for whatever reason. Then the study will conclude that butter is bad for people.
Another study, with a different group of people, may have accidentally included people whose reactions to butter are all over the spectrum. This study will conclude that butter is irrelevant to health.
And a third group may have an unusual number of people who respond well to butter, or who respond badly to margarine. This study, then, will conclude that margarine is bad, and butter is good.
All of these studies would miss a fundamental issue: the importance of genetic differences. And that will invalidate their findings. Worse, from a layperson’s point of view, it gives contradictory signals, and erodes confidence in medical research.
The Implications of Genetic Testing
Genetic testing is still in its infancy, so that even though it has already taken enormous strides, there is still a tremendous amount we can and eventually will learn from it. It will take time for research to come up with clear indications of how particular foods interact with an individual’s genetics, but eventually you’ll be able to get an analysis of your DNA (for less than, say, $100), that will tell you how different foods will affect you.
Some foods (like leafy, green vegetables) will be great for you, and are ones you should eat regularly. This will be your “A” list. Some foods (like fish, chicken, and whole grains) will be good for you, but should be eaten in moderation (the “B” list). Some foods, like hot fudge sundaes, red meats, and white breads, you can eat occasionally, but shouldn’t eat often – this is your “C” list. And some foods may be positively harmful to you, say if you have an allergy to nuts or dairy products, or a food sensitivity to things like wheat gluten or corn. This is the “X” list, and you should always avoid everything on it.
Each list will be different for each person, although there will be large areas of overlap. I doubt, for instance, whether hot fudge sundaes will be on anyone’s “A” or “B” lists. I suspect that genetic variations will probably show the greatest differences in each person’s “X” list, which spells headaches for foodservice organizations, which already have long lists of allergies of which they need to be aware.
The same will apply to other environmental factors, such as the adhesives used in laying carpet, or particular kinds of trees, flowers, or aromas. This may eventually change how you decorate your house, or even which materials you use to build it.
As a result, genetic analysis will affect a broad range of industries, from farming through foodservice, to builders and construction companies, to schools and public institutions, and many more. In fact, I suspect that the more we learn about our genomes and how they work, the more aware we will be of things that have affected us, perhaps without us ever knowing about them.
Imagine, for instance, that we didn’t know that some people are allergic to ragweed or pollen. We’d be mystified by their symptoms. It’s this kind of revelation that we will consistently stumble upon as we learn more about how our bodies (or, more accurately, our biomes) interact with our environment. It will be both revolutionary, as new discoveries lead to blazing insights, and evolutionary, as we learn how to make adjustments in diagnosis and treatment for ever-smaller groups of individuals.
But one industry is already being shaken to its core: pharmaceuticals.
What’s Ahead for Big Pharma
We already know, from drugs like Herceptin, that genetic testing can make an enormous difference in the value and effectiveness of a drug. Herceptin can be very effective for certain kinds of breast cancer – but only if the patient has a particular genetic pattern. Otherwise, the only thing Herceptin will do for them is to empty their wallets.
The ironic part about Herceptin is that it was originally slated to be dropped because it wasn’t effective enough across a broad spectrum of patients. It was developed by drug company Genentech in cooperation with the University of California at Los Angeles (UCLA). Yet, once prospective patients were screened for a particular receptor, the results were remarkably improved. Hence, a drug that was going to contribute only to the cost of failed drug research wound up being a big money maker for Genentech and UCLA.
But until very recently, Herceptin was largely an exception. There are a few, but only a few, other stories like it. And this is largely because genetic screening didn’t fit the business model of Big Pharma.
Until they accepted the inevitable, major drug companies needed to sell millions of doses for billions of dollars with big, blockbuster drugs. They more or less had to do this because they are so big that smaller drugs wouldn’t move the needle on their revenues and profits. Without big increments in income, Wall Street, with its miniscule attention span, gets restless about drug company results, and hence, with drug company managements.
The problem, therefore, with genetic screening is that it can take a drug that either isn’t very effective, or has unacceptable side effects – the two biggest reasons why drugs are dropped after years of development – and turn it into a modest success. Such drugs, once given only to the appropriately screened patients, will either prove to be very effective, or not have anywhere near the side effects that it would have on a broader group of patients.
But Wall Street isn’t interested in modest successes. The problem is that the screened drugs will sell to a much smaller potential market, with tens of thousands of doses bringing in tens or hundreds of millions of dollars in revenue – which isn’t enough. Yet, despite this, some drug companies are changing as they get into synch with the broader underlying knowledge of genetics.
Why New Drugs Are So Costly
I believe Big Pharma was ignoring a real opportunity, because revenue is only half of the profit equation.
Coming up with a figure of how much it costs to develop a successful new drug from scratch is difficult and controversial. I’ve seen numbers that range from $800 million to $1.2 billion per drug, and those estimates are several years old. These figures are controversial partly because the drug companies have a vested interest in pumping them up as a means of defending the high prices they charge for new drugs. Hence, a big number is politically more useful than a smaller one.
Yet, no matter who does the counting or what their motives are, it is still true that it’s very expensive to develop a new drug. And one of the biggest expenses in drug development is discarding drugs that seem promising, but don’t prove out for some reason.
Before approval, drugs go through several stages, from lab tests, to animal trials, to four stages of human trials. Each stage of the development and approval process is significantly more expensive than the ones before it. Yet, a drug that fails in human clinical trial Stages III or IV has incurred substantially all of the costs of drug discovery, and usually fails because it is either not as effective with a broad spectrum of people as earlier stages indicated, or because unacceptable side effects (up to, and sometimes including death) are more prevalent than expected.
So drugs that get to Stage III or IV clinical trials have shown that they have a high probability of being valuable. If they fail at that point, they represent the highest costs in the entire drug development process.
Why Lower Sales May Be More Profitable
As the drug companies start to do genetic screening while drugs are being started in human trails, they will inevitably rescue many of the drugs that fail late in the process – and save themselves the huge development costs involved. This would mean that although their revenues per drug might fall by a significant amount, so, too would their costs. In other words, their revenues might go down, but their profits could go up.
That’s not the model that Big Pharma and Wall Street have liked in the past – but increasingly that’s the way that drug development is going. And research results could be improved further by using novel data searching techniques, such as evolutionary algorithms, to further lower development costs.
But first more drug companies have to accept that the old business model is dying, and that the tools that are now emerging should be used to create a better, more profitable model.
And, as our understanding of how our individual genetics interact with our environment, at the very least we’ll find out whether we should be eating butter or not.
© Copyright, IF Research, September 2015.
