Modh Karim, a first year DPhil in Population Health
Heart disease, stroke, cancer and diabetes – it is difficult to find someone who has not had a friend, relative, or family member afflicted by one of these scourges. With the recent advent of an array of diagnostic tests and novel drugs, we have made remarkable strides in preventing premature death but not nearly enough to pat ourselves on the back. Our treatments often begin when the disease has already progressed to an advanced stage; what went on to ignite the first event inside our body that set off the disease – we have very little clue.
Inflammation is the body’s response to foreign agents like bacteria and viruses. It is also what makes you scratch the itch of a mosquito bite, sneeze/cough when you’re coming down with a cold or when running up a fever. The system is designed to unleash an infantry of chemicals designed to kill all things alien, protecting you in the process. But there’s a different kind of inflammatory response happening inside your blood vessels. We believe this inflammatory process is central to formation of clots or bleeds in your heart and brain, diabetes-induced nerve and vessel damage and even cancer in many cases. Unfortunately, this happens with a completely different set of chemical mediators than the one simply causing a rash and we have yet to figure which one!
For years, it was believed CRP (C-reactive protein) – a molecule produced from your liver as part of the inflammatory cascade – is responsible for all this havoc. However, the evidence implicating CRP were all based on observational studies – a type of experiment where people are only observed at a specific time with or without follow-up over a period of time. These studies are fraught with biases. For example, one very large observational study showed CRP was associated with increased risk of heart disease, stroke and even death from several cancers.[1] So would it now be appropriate to invest our limited resources on a drug that blocks the action/production of CRP? Well, not quite and at least not yet. That large study combined data from a series of smaller observational studies and most of these studies have either measured CRP after the disease (resulting in the so-called reverse causation bias) or did not take in account of all the other things that could distort the relationship of CRP and chronic diseases (called confounders), for example – smoking or lack of physical activity. Hence, it was difficult to be reasonably certain CRP was responsible for the observations in these studies.
How do we go about solving this mess? The unequivocal answer: an intervention study – carrying out a randomised controlled trial (RCT), which (in a simple scenario) randomly allocates study participants in to two groups – one which provides them with a drug that lowers their CRP levels and the other gives the participants placebo or a dummy pill which does nothing. They are followed up periodically to assess their CRP levels and disease status. Assuming this allocation was completely at random, we would expect the factors (both known and unknown) or confounders, which would have otherwise misled us like observational studies, to be evenly distributed in both groups. Since the trial allocated participants before they had any disease, we would also overcome the reverse causation bias. If the participants were clueless (blinded) to whether they were on the active or dummy pill, we can safely assume their reports in follow-up interviews (for example, on any heart disease symptoms) should be reasonably valid.
But here are the caveats. RCTs are expensive. Drug industries go through a series of such RCTs to license the use of a drug for a specific indication. Most drugs fail in the process but not before costing the pharmaceutical industry billions of pounds and the rare ones which are ready for the market are priced beyond reason. One year’s worth of Herceptin – a drug for breast cancer – costs the NHS more than £20,000 per person! Further, the trials run for a short period of time and may fail to capture all kinds of adverse effects. For some drugs of questionable efficacy, you wouldn’t even get the ethical clearance to test in populations who may need it the most such as pregnant women.
However, there is a potential solution and it has become the crux of recent research. At the start of conception, when the sperm fuses with the egg to start the process of creating a new life – part of the process distributes future traits (like genes for high or low CRP levels) of the potential person completely at random. So, at the population level, you would expect some to have genetically high, normal and low CRP levels – again, completely at random. Nature, it appears, has done the hard job of random allocation for us; some go as far to call this nature’s own RCT![2]
It is now possible to use these genes as instruments to mimic effects of a drug (e.g., lowering CRP). How? We can compare individuals, genetically predisposed (and hence, lifelong) to produce lower CRP concentrations, to their peers producing elevated levels of CRP and see if it really makes a difference, i.e., whether the incidence of stroke, for example, is significantly higher in those with genetically higher CRP levels than those with normal or low CRP levels.
My research uses this approach to answer whether drugs reducing levels of inflammatory chemicals (like CRP) are really worth the money and time of the pharmaceutical industry. In addition to prioritising drug targets, this approach – dubbed Mendelian randomisation – assesses the lifetime impact (as opposed to the short RCT runs) of higher or lower levels of a biomarker (provided we know the genes encoding it) on heart disease, stroke, diabetes and cancer. This approach can also yield better insight into the potential adverse effects of drugs and who are likely to have them based on individual genetic make-up. In the long run, it can help us steer clear through the noise, make the right choices and potentially save millions of lives (and billions in taxpayer money)!
[1] Emerging Risk Factors C, Kaptoge S, Di Angelantonio E, Lowe G, Pepys MB, Thompson SG, et al, ‘C-Reactive Protein Concentration and Risk of Coronary Heart Disease, Stroke, and Mortality: an Individual Participant Meta-Analysis’, Lancet, 2010; 375(9709):132-40.
[2] Thanassoulis G, O’Donnell CJ, ‘Mendelian Randomization: Nature’s Randomized Trial in the Post-Genome Era’, JAMA, 2009; 301(22): 2386-8.