This is poorly formatted because I ripped it out of a pdf I wrote a while back for which I can't find the original, but it should still be readable.
Science and Drugs Workshop - for Edinburgh University Debates Union 2016-17
Scientific Aims and Progress
No consensus over how science works – when people talk about ‘the scientific method’, they either don’t know what they’re talking about or they’re referring to a broad set of methods which they think of as constituting science. This means that what science is is always up for grabs.
What science isn’t: truth; objective; value-free; unbiased; cheap.
This is pretty much only important in debating insofar as it allows us to (a) talk about
research incentives and (b) criticise debates which talk about replacing some policy- making process with a ‘panel of experts’ and inevitably end up with the prop team claiming that this will work perfectly and be lovely.
If you want a rough and ready definition of science, you can always go for “a process by which we try to predict and explain the world around us”.
Research & Funding
About two thirds of scientific and technical research is performed by
corporations and industry. About twenty percent is performed by universities, ten percent by governments, and the remaining few percent are split between e.g. charities, philanthropic endeavours, crowdfunding.
“Basic” research (i.e. research which doesn’t have immediate applications, whether they be commercial, military, technical, etc) is more often funded by states – only the most research-oriented of corporations tend to fund blue- sky research rather than near-term focussed R&D. These include IBM (e.g. Watson and cognitive computing), L’Oreal (have an award for women in science), SpaceX, Google.
Common critiques of basic research include the idea that it is a waste of taxpayers’ money to fund a scientist’s curiosity into the shape of hedgehogs’ ears. However, basic research has often expanded our understanding of the universe in unlikely ways, as well as leading indirectly to important concrete applications.
There isn’t always a hard and fast distinction between basic and applied research. When people say that scientists have no moral responsibility with respect to their research because they only perform it, and other people use it for good or ill, they’re (probably) wrong. This is partly because the choice to fund a particular kind of research is value-laden, but also because the doing of that research may itself necessitate an application. Good example is the Trinity Test, which demonstrated the possibility of nuclear fission by... doing nuclear fission. The application couldn’t be separated from the
One exception to the “state funds basic research” rule is in biotechnology,
which is dominated by industry. You’ve probably heard of Monsanto, who are the really big company involved in GMOs etc. However, there’s a pretty large start-up and amateur culture surrounding biotechnology, including in areas like synthetic biology.
Crispr-Cas9, which is a technology used to edit gene sequences and is currently in the process of revolutionising a lot of biological research, is cheap enough and easy enough to use that it can basically be utilised by geeks in their basements. This is very similar to the kind of home programming seen in the 1980s before the rise of big operating systems like Windows.
Theranos, an ill-fated start-up promising to change the way in which blood testing is performed, is a good example of how biotechnology (particularly in Silicon Valley) is still a bit of a Wild West. It was founded by Elizabeth Holmes, aged 19, in 2003, and by 2014 it managed to raise $400m in funding and had an estimated value of $9bn. It was revealed by an investigative journalist in 2015 that they hadn’t actually been using their own patented technology most of the time, and that it didn’t in fact really work. Holmes now has an estimated worth of $0, and the company has sunk to $800m and has closed most of its lab operations. This raises some issues with respect to the secrecy of business ventures, as they refused to release any of the details of how their systems supposedly ran into the academic community. This reveals tensions between the “free inquiry” basis of scientific endeavour, and the profit motives of privately owned corporations whose interests are in keeping their trade secrets, secret.
There are also huge controversies over the patenting of genes and other DNA sequences. Association for Molecular Pathology v. Myriad Genetics, inc, is the go-to landmark legal case. Myriad had a product called BRACAnalysis, which detects particular mutations in the BRCA1 and BRCA2 genes that put women at high risk for both breast and ovarian cancer. Their business model was to exclusively offer these diagnostic services, with investors putting money into the company on the basis that the 20-year life of the patent on the diagnostic testing would allow them to recoup their investment, and it was this money that allowed them to rapidly sequence the BRCA2 gene and create a robust diagnostic test. To make this model workable, they would have had to enforce their patents against competitors, including university diagnostic labs. TL;DR – they patented particular sequences of DNA. The Supreme Court eventually ruled that naturally-occurring DNA was not patentable, but cDNA was because it did not occur naturally. This makes it difficult for companies such as Myriad, because previous USSC decisions have rendered most diagnostic tests unpatentable (In re Bilski and Mayo v. Prometheus, if you’re interested) and so patenting DNA sequences was one of the only ways that competition could be staved off. Lingering questions over whether it’s ok to patent gene sequences.
a. Drug research is expensive. Really, really expensive. Last time I checked, it costs around $500m and 10-15 years to bring a drug from design to market. This is because in 1962, it was found that the drug thalidomide, marketed to
pregnant women as a way of preventing morning sickness, was a teratogenic drug which resulted in babies being born with malformed limbs. Previously, every new drug was seen as beneficial. From then on, drugs were subjected to a great deal more scrutiny.
b. Development steps:
Search for a target. A drug needs something to act on. You want a
protein or a gene.
Look for a compound which might affect the target. Huge compound
libraries are developed and kept, and assays (processes which quantify the interaction between a compound and a target) are used to test the compounds, thousands at a time, through a process called high-throughput screening. Tiny changes can be made to molecules to see if this makes a difference to the efficacy of the compound.
Test for pharmacokinetic behaviour in animals: how is it absorbed in the gastrointestinal tract, how is it distributed throughout the body, how is it metabolised and excreted? It’s all well and good having a compound which works on your target gene but if it kills everything it touches then it’s not going to be a popular product.
Preclinical studies used to evaluate the drug’s safety, efficacy and toxicity in animal models. They’re also used to test that the drug isn’t carcinogenic, mutagenic or teratogenic. Usually drugs must be tested on both a primate and non-primate animal in order to effectively model how they might act in humans.
Phase I clinical trial. Testing safety. A small number of paid healthy volunteers (20-100) are given the drug to determine the drug’s safety, the safe dosage range, and any side effects.
Phase II clinical trial. Testing efficacy. Here the drug is tested for the first time in patients with the target disease or condition. These last from several months to years. They help determine correct dosage, common short-term side effects and the regimen to use in larger clinical trials.
Phase III clinical trial. Testing the drug’s benefit in a large targeted patient population with the disease. Confirm efficacy, monitor side effects, compare the drug to other commonly used alternative treatments. Used to gain info on the overall risk-benefit profile of the drug. These trials should be double-blinded if possible: neither the doctor nor the patient should know whether they are receiving the drug or a placebo/alternative drug. They take place over several years, at multiple locations around the world (a subject of controversy where they are tested on patients who could never afford the drug in the developing world, because to be a subject of a clinical trial is to be subject to potential risk, and the common justification for this risk is that you, or people like you, might benefit from the drug in the future).
Phase IV clinical trials. Marketing and safety monitoring. These studies are conducted (in the US) after FDA approval. They are often used primarily for medical marketing. They can also be used to generate information about the drug’s long term safety, benefits and side effects. They can also be used to test the drug in different patient populations (e.g. children), in new delivery modes, or for uses on different conditions.
c. The upshot of this long and arduous process is that only very few institutions have the capital required to be able to bring new drugs to market. States don’t want to take the risk that a drug won’t work (because wasting taxpayers’ money is seen as a cardinal sin), and so pretty much only private companies are willing to front the time and money required.
This means there are extremely high barriers to entry into the pharmaceutical industry. Start-ups exist, but they’re not that common and they usually get swallowed by one of the big pharma corporations (e.g. Burroughs-Wellcome, GlaxoSmithKline, Merck, Pfizer, Roche, AstraZeneca, Bayer, Eli Lilly).
As such, research into pharmaceuticals tends to prioritise what would be profitable, rather than necessarily what would provide the greatest benefit to humanity. This results in the neglect of a large number of diseases which kill many people: in contrast with the “Big Three” (HIV/AIDS, malaria, tuberculosis) which receive a large amount of funding, the “Neglected Tropical Diseases” including Dengue Fever (50-100 million infections annually), Chagas Disease (15 million infected, debilitating chronic symptoms), and Leishmaniasis (12 million infected, 20,000 deaths per year) receive comparatively little. These diseases are often treatable only with expensive medication.
You’ll probably also have heard of antibiotic resistance, and the fact that it’s likely to really start hurting us within the next decade. Some people argue that pharmaceutical companies have no incentive to research new antibiotics (the last new class was discovered around the 1960s/70s) because they’re not profitable as they’re only prescribed for short courses. Whilst this might be true to an extent, it’s by no means the whole story. Much of the reason we’ve found it so hard to create new antibiotics is because we’ve exhausted most of the low-hanging fruit in terms of naturally occurring kinds. We’re having to look increasingly further afield to find new ones: right now there’s research going on with e.g. Komodo Dragon saliva which hopes to find novel antibiotics in around exotic bacteria.
d. A couple of miscellaneous ethical issues:
i. The people who participate in Phase I clinical trials are often students, homeless, or otherwise financially insecure individuals. Questions over whether they can truly give informed consent into the procedures, given they often need the money. Some individuals are professional ‘guinea pigs’, doing a number of different Phase I trials every year in order to stay alive. More questions over whether they can meaningfully consent if they can’t opt out (in many instances those who do so are not paid in full, even though legally they must be). They’re treated as volunteers or contractors, although in many instances it’s more like a job.
The justification for individuals undergoing harm in clinical trials is normally that they, or people like them, will benefit in the future from that harm. This, at least, is the justification for Phase II and III trials, as participants in these trials are not paid and have a 50% chance of being given a drug that is not the one being tested (and even if they get the trial drug, it might be an inferior treatment). Given that tests are increasingly outsourced to the developing world because of laxer regulations and lower costs, it’s questionable how well this justification holds up, as most of the individuals being tested are unlikely to ever be able to afford the drug.
This also has epistemic issues attached. If you are testing your drug on one population and using it in another, then there’s a non-trivial possibility it will act differently in your other population, especially if they have different diets, lifestyles, genetic propensities towards particular conditions etc. If you’re testing your drugs on poor people, in the developing world and in your own country, there’s a strong chance that the treatment population is quite medically different
from the test population. This is the opposite problem to that found in social sciences, where the vast majority of test subjects are “WEIRD” (Western, educated, from industrialised, rich, developed countries) and there are also questions over generalisability.
One further epistemic issue: pharmaceutical trials which are funded by pharma companies return results which are far more likely to be positive than those which are funded by a comparatively neutral body. And even when they’re not positive, these results tend to get buried. There is currently a campaign to have all drug trial data made publicly available, which would allow for much greater scrutiny of research practices.
In the real world, money is the main currency. In the world of academia, your publications and your reputation as a researcher form a surrogate currency, and it’s through these that you gain access to money (in the form of research grants, consultancy fees, jobs with better pay, etc). This means that there is a pressure to publish your research in peer-reviewed journals, and to do so often and in as prestigious journals as you can.
High quality journals receive far more article submissions than they could ever publish. This means that they tend to publish those which are likely to have the largest impact. This creates some issues: the studies which are most likely to have a big impact are often the hardest to replicate (part of the “scientific method” used to verify results); they’re incredibly unlikely to publish studies which have negative results or which are replications; you’re less likely to get published if you’re not from a prestigious university, etc.
What this means is that there are somewhat toxic incentives in academia. There is a pressure to find significant results (even if they aren’t necessarily there), to publish as many papers as possible (to the detriment of quality), to perform novel studies rather than verifying existing data. In order to advance your career, you have an incentive to perform work that isn’t necessarily of a high quality.
There’s also a strong emphasis on networking in academia. This tends to
disadvantage people from non-traditional backgrounds because of all the Old Boys’
Club nonsense that we’ve seen play out in every other field in life.
Most big journals are owned by a small number of companies (Elsevier, SAGE,
Wiley). Access to them is incredibly expensive. You’re looking at $50 for a single article, or hundreds of dollars for a subscription. This means that access is basically unaffordable for private individuals, limiting the ability of people outside of universities to get access to esoteric knowledge. Universities buy access to the big journals, but Elsevier et al will usually package up this access with much smaller, less popular journals which they also have to pay for, meaning universities spend an awful lot on journals which nobody really uses.
This is one of the reasons for the huge expansion of Open Access Journals. These journals are free to read (which is nice), but there are still issues. The first is the question of quality: there have a number of incidents in recent years of poor quality papers being accepted for OA Journals (such as one which simply read “Get Me Off Your Fucking Mailing List” over and over, and another paper for a Nuclear Physics conference which was created entirely in iOS autocomplete). Then there’s the money: OA Journals are “pay-to-play”: the academics who submit papers have to pay to get them published.
Lots of people think that science should be open, and scientific knowledge should be free. Sci-Hub is a Russian website which has pirated millions of papers and made them freely available to anyone. Its mission statement is “to remove all barriers in the way of science”. Similarly, Aaron Swartz was a prominent programmer involved in the construction of RSS, Reddit, Creative Commons and a number of other important projects. He took his own life in 2013 after he was arrested over his attempt to download millions of articles from JSTOR. Access to information is deeply political.
When a pharmaceutical company first develops a new drug to be used for a disease condition, it is initially sold under a brand name by which the clinicians can prescribe the drug for use by patients. The drug is covered under patent protection, which means that only the pharmaceutical company that holds the patent is allowed to manufacture, market the drug and eventually make profit from it.
In most cases in the US, the patent is awarded for 20 years, but the lifetime of a patent varies between countries and drugs. The patent is applied for long before clinical trials, which means that the effective patent period after the drug has been approved is normally around seven to twelve years.
Once a patent has expired, the drug can be manufactured and sold by other countries as a “generic”. These drugs are identical to the branded drug in terms of their efficacy, safety, usage, pharmacokinetics, route of drug administration and pharmacodynamics.
A drug can be manufactured as a generic when the following apply:
a. Its patent has expired
The company that would manufacture the generic drug certifies that the patents held on the drug are unenforceable, are invalid or would not be infringed upon
There have never been any patents on the drug before
In countries where there is no patent protection for the drug (this opens up
some drugs to cheaper (and potentially poorly regulated) manufacture in
other countries, from where they can be illegally exported elsewhere).
5. Impacts for the consumer:
Once the generic drug is on the market, the monopoly of the patent holder is removed.
This encourages competition and results in a significant drop in drug costs, which ensures that life-saving and important drugs reach the population at competitive prices.
6. Some things worth bearing in mind:
If companies are able to find another indication for a drug (i.e. another condition for which it can be used), they can often extend their patent life or re-patent it for this kind of usage. This leads to pharmaceutical companies encouraging the codification of new conditions (Generalised Anxiety Disorder and Pre-Menstrual Dysphoric Disorder are two good examples) for which an existing drug can be prescribed.
Huge pharmaceutical companies often make the bulk of their profit (which they can then plough back into R&D) from one or two drugs (e.g. Pfizer manufactures Viagra). They initially rely on the patent to protect them, but once this has expired they are open to competition. One way of circumventing this competition is to exploit the ignorance of the general public as to the lack of difference between branded and generic drugs. This is why people still buy Viagra, rather than generic sildenafil citrate. Marketing is hugely important for this: Nurofen is just ibuprofen, but people will still pay 10x as much for it as they would for generic ibuprofen.
Once a drug is approved for prescription for one condition, it can be prescribed “off-label” for other conditions. This is at the discretion of doctors, but they are not required to tell patients that the drug they are being given is off-label. This usage does not require testing for efficacy or safety over and above the trials which have already been completed. Off label prescribing has benefits when patients have exhausted pre-approved options, as may be the case for rare diseases or cancer. It does raise the potential for lawsuits if off- label usage has unwanted side effects.
i. A good example of an off-label usage scandal is Fen-Phen. The FDA approved medications fenfluramine hydrochloride and phentermine hydrochloride as individual, short-term treatments for obesity. However, doctors eventually began prescribing the two drugs together after an article which described dramatic effects for weight loss of this cocktail appeared in a medical journal and mainstream publications. Many patients ended up with severe, and potentially deadly, heart-valve damage. This triggered a multi billion dollar lawsuit. Fen-Phen was ordered off the market in 1997.